ETC PSB2163P(V3.1)

ICs for Communications
Audio Ringing Codec Filter
Featuring Speakerphone Function
ARCOFI®-SP
PSB 2163
User’s Manual
ICs for Communications
Audio Ringing Codec Filter
Featuring Speakerphone Function
ARCOFI®-SP
PSB 2163
User’s Manual 06.95
PSB 2163
Revision History:
Current Version 06.95
Previous Version:
02.94
Page
Subjects (changes since last revision)
108
Absolute Maximum Ratings
108
DC Characteristics
110
Analog Front End Characteristics
111
Transmission Characteristics
Test Conditions corrected
Addition of Overall Programming Range
112
IOM®-2 Bus Timing
Specification of Jitter Timing
Data Classification
Maximum Ratings
Maximum ratings are absolute ratings; exceeding only one of these values may cause
irreversible damage to the integrated circuit.
Characteristics
The listed characteristics are ensured over the operating range of the integrated circuit.
Typical characteristics specify mean values expected over the production spread. If not
otherwise specified, typical characteristics apply at TA = 25 °C and the given supply
voltage.
Operating Range
In the operating range the functions given in the circuit description are fulfilled.
For detailed technical information about “Processing Guidelines” and “Quality
Assurance” for ICs, see our “Product Overview”.
General Information
Contents
Page
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Comparison between PSB 2163 and PSB 2165 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Table of Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1
1.1
1.2
1.3
1.4
1.4.1
1.4.2
1.4.3
1.4.4
1.4.5
1.4.6
1.4.7
1.4.8
Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pin Definitions and Functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Logic Symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Functional Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
System Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ISDN-Voice Terminal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Terminal Adapter a, b for Analog Telephones . . . . . . . . . . . . . . . . . . . . . . . .
Voice/Data Terminal (PC-Card) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multifunctional ISDN-Terminal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Digital Voice Terminal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IOM®-2 Line Card Application. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Primary Rate Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Group 3 Fax / Modem Adapter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
13
16
17
18
19
22
22
23
25
27
28
29
2
2.1
2.1.1
2.1.2
2.1.3
2.2
2.2.1
2.2.2
2.2.3
2.2.4
2.2.4.1
2.2.4.2
2.2.4.3
2.2.4.4
2.2.4.5
2.2.4.6
2.2.5
2.2.5.1
2.2.5.2
2.2.5.3
2.2.5.4
2.2.5.5
2.2.5.6
2.2.5.7
2.2.6
Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analog Front End (AFE) Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Description of the Analog I/O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AFE-Attenuation Plan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interface to Acoustic Transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ARCOFI® Signal Processor (ASP) Description . . . . . . . . . . . . . . . . . . . . . . .
Transmit Signal Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Receive Signal Processing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Programmable Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tone Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tone Generation Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Control Generator. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tone Generator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tone Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tone Level Adjustment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DTMF-Generator (transmit) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ARCOFI® Speakerphone Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Speech Detector. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Speech Comparators (SC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Attenuation Control Unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Speakerphone Test Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Automatic Gain Control of the Transmit Direction (AGCX) . . . . . . . . . . . . . .
Automatic Gain Control of the Receive Direction (AGCR) . . . . . . . . . . . . . . .
Loudhearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Speakerphone Coefficient Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
30
31
32
32
34
35
37
37
39
40
40
42
43
46
47
47
48
50
52
58
59
59
61
62
63
Semiconductor Group
5
General Information
Contents (cont’d)
Page
2.3
2.3.1
2.3.2
2.3.3
2.3.4
2.4
ARCOFI Digital Interface (ADI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PCI-Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IOM®-2 Frame Structure and Timing Modes . . . . . . . . . . . . . . . . . . . . . . . . .
Serial Control Interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Serial Data Interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Test Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
65
65
65
69
71
72
3
3.1
3.2
3.3
3.4
3.4.1
3.4.2
3.4.2.1
3.4.2.2
3.4.2.3
3.4.3
3.5
Operational Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Initialization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ARCOFI® Operating Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IOM®-2 Interface Protocol. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B- and IC-Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Monitor Channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
MON-Channel Data Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
MON Transfer Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Implementation of the MON-Channel Protocol. . . . . . . . . . . . . . . . . . . . . . . .
Command/Indication Channel 1 (TE-mode). . . . . . . . . . . . . . . . . . . . . . . . . .
ARCOFI® Voice/Data Manipulation (VDM). . . . . . . . . . . . . . . . . . . . . . . . . . .
73
73
74
79
81
81
81
81
83
85
87
89
4
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
4.12
4.13
Detailed Register Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Command Register (CMDR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
General Configuration Register (GCR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Data Format and Interface Configuration Register (DFICR) . . . . . . . . . . . . . 97
Programmable Filter Configuration Register (PFCR). . . . . . . . . . . . . . . . . . . 98
Tone Generator Configuration Register (TGCR) . . . . . . . . . . . . . . . . . . . . . . 99
Tone Generator Switch Register (TGSR). . . . . . . . . . . . . . . . . . . . . . . . . . . 100
AFE Transmit Configuration Register (ATCR) . . . . . . . . . . . . . . . . . . . . . . . 101
AFE Receive Configuration Register (ARCR) . . . . . . . . . . . . . . . . . . . . . . . 102
Test Function Configuration Register (TFCR) . . . . . . . . . . . . . . . . . . . . . . . 103
SDI Configuration Register (SDICR); SDI-mode only . . . . . . . . . . . . . . . . . 104
Time-Slot Configuration Register (TSCR); SDI-mode only . . . . . . . . . . . . . 105
Extended Configuration Register (XCR) . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Test Mode Register (TMR). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
5
Electrical Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
6
Package Outlines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
7
Application Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
8
Information on Literature. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
Semiconductor Group - Addresses
9
Microelectronic Training Center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
IOM®, IOM®-2, ARCOFI®, ISAC®-P, ISAC®-S, ITAC®, SICOFI® are registered trademarks of Siemens AG.
Semiconductor Group
6
General Information
Introduction
The PSB 2163 ARCOFI®-SP provides the subscriber with an optimized Audio, Ringing,
Codec, Filter processor solution for a digital telephone. It fulfills all the necessary
requirements for the completion of a low-cost digital telephone.
Please note:
Throughout this whole document “ARCOFI®” refers to ARCOFI®-SP
PSB 2163.
The ARCOFI performs all coding, decoding and filtering functions according to the
CCITT and ETSI (NET33) norms.
Full featured applications are possible without any external elements. All the necessary
hardware and software is implemented. In addition the ARCOFI offers a speakerphone
and monitoring function. This feature is completely digitally implemented in the chip.
Two transducer correction filters (one for each direction) can be programmed to correct
the analog transducer frequency characteristics.
The ARCOFI provides a universal DTMF, tone and ringing generator for the receive
direction. The signal forms available are DTMF, square, trapezoid and sine wave.
Complex signal sequences are made possible by a control generator (e.g. pulsed three
tone call in conjunction with the beat generator).
A DTMF-generator for the transmit direction is also available. If the transmit
DTMF-generator is active, only a part of the receive tone generator function is possible.
This flexible tone generator concept fulfills a wide range of applications.
The interfacing to a handset mouth and earpiece is facilitated by a flexible analog front
end. A loudspeaker output has also been integrated on the chip as well as a secondary
input for a handsfree microphone. All analog inputs and outputs are gain programmable
through software.
At the digital side an ISDN-Oriented Modular (IOM®-2) interface for terminal (TE) and
non-terminal (non-TE) applications or a Serial Control/Data Interface (SCI & SDI) is
realized to connect layer-1/2 devices to the ARCOFI.
The ARCOFI is a BlCMOS-device, available in a P-DIP-28, P-LCC-28-R or P-DSO-28
package. It operates from a single + 5 V supply and features a power-down state with
very low power consumption.
Semiconductor Group
7
General Information
Comparison between PSB 2163 and PSB 2165
Table of main differences:
PSB 2163
PSB 2165
1-µ BICMOS technology
2-µ CMOS technology
Comment
P-DSO-28
SLD-interface
Non-TE IOM-2 interface
Serial Control/Data Interface
One receive channel
Two receive channels
Digital high-pass in receive
direction
AGC in receive direction
LGA (programmable gain
stage) in receive direction
Sidetone gain stage GZ with
higher resolution (two byte
coefficient)
GZ with one byte coefficient
Optimized speakerphone
function
Extended coefficient
space is necessary
Microphone amplifier with
additional 36-dB and 42-dB
gain stages
Controlled monitoring with
fixed attenuation
Controlled monitoring with
programmable attenuation
Ringing directly via
Ringing via loudspeaker over
loudspeaker
the second receive path
with square-wave in 3-dB steps
Tone generation unit can be
switched to transmit path and
added to transmit speech
signals
Additional test loops
CCITT G.714
ETSI (NET33) & CCITT G.714
Semiconductor Group
8
General Information
Table of Symbols
AD
A/D
ADI
AFE
AGCX
AGCR
AHO
AIMX
ALC
ALF
ALI
ALN
ALS
ALTF
ALZ
AMI
ARCOFI
ASP
Address of the ARCOFI (IOM-2 mode)
Analog to Digital converter
ARCOFI Digital Interface
Analog Front End
Automatic Gain Control Transmit
Automatic Gain Control Receive
Handset Output Amplifier
Analog Input Multiplexer control bits (ATCR)
Analog Loop via Converter (TFCR)
Analog Loop via Front End (TFCR)
Analog Loop via Interface (TFCR)
Analog Loop via Noise Shaper (TFCR)
Loudspeaker Amplifier
Analog Loop & Test Function bits (TFCR)
Analog Loop via Z-side tone gain stage
Microphone Amplifier
Audio Ringing Codec Filter
ARCOFI Signal Processor
BM
BT
Beat Mode bit (TGCR)
Beat Tone bit (TGCR)
CAM
CCITT
CG
CMDR
COP
CR
CRAM
CS
Chip Address Mode bit (IOM-2 two chip mode; GCR)
International Telegraph and Telephone Consultative Committee
Control Generator bit (TGCR)
Command Register
Coefficient Operation (CMDR)
Configuration Register
Coefficient RAM
Chip Select active low (serial control interface)
D/A
DCE
DCL
DCLK
DD
DEC
DHON
DHOP
DHPR
DHPX
DLN
DLP
Digital to Analog converter
Double Clock Enable at DCLK pin (SDICR)
IOM-2 interface clock
Data Clock pin (serial data interface)
IOM-2 Data Downstream pin
Decimation filter
Disable pin HON (XCR)
Disable pin HOP (XCR)
Disable High Pass Receive bit (PFCR)
Disable High Pass Transmit bit (PFCR)
Digital Loop via Noise Shaper (TFCR)
Digital Loop via PCM-register (TFCR)
Semiconductor Group
9
General Information
Table of Symbols (cont’d)
DLS
DLSN
DLSP
DLTF
DR
DRAM
DSP
DT
DTMF
DU
DX
Digital Loop via Signal processor (TFCR)
Disable pin LSN (XCR)
Disable pin LSP (XCR)
Digital Loop & Test Function bits (TFCR)
Data Receive pin (serial data interface)
Data RAM
Digital Signal Processor
Dual Tone bit (TGCR)
Dual Tone Multi Frequency bit (TGSR)
IOM-2 Data Upstream pin
Data Transmit pin (serial data interface)
EP0
EPP0
EPP1
EPZST
ETF
ETSI
EVX
EVREF
EWDF
Earpiece
Enable Push-Pull at pin DU/DX (SDICR)
Enable Push-Pull at pin SA/SDX (SDICR)
Enable PZ1/PZ2 to output internal Status conditions (TFCR)
Enable Tone Filter bit (TGCR)
European Telecommunications Standards Institute
Enable Voice Transmit bit (GCR)
Enable VREF buffer bit (ATCR)
Electrical Wave Digital Filter
FR
FSC
FX
Frequency correction Receive bit (PFCR)
IOM-2 and SDI-Frame Synchronization pin (8 kHz)
Frequency correction Transmit bit (PFCR)
GR
GX
GZ
Receive Gain bit (PFCR); Receive gain stage
Transmit Gain bit (PFCR); Transmit gain stage
Z-side tone Gain bit (PFCR); Z-side tone Gain stage
HO
HOC
HON
HOP
Handset Output
Handset Output Control bits (ARCR)
Handset earpiece Output – pin
Handset earpiece Output + pin
IDENT
IDR
INT
IOM
ISDN
Identification Code
Initialize Data RAM (TFCR)
Interpolation filter
ISDN-Oriented Modular
Integrated Services Digital Network
Semiconductor Group
10
General Information
Table of Symbols (cont’d)
LAW
LIN
LS
LSC
LSN
LSP
A-Law/µ-Law bit (GCR)
Linear data mode (VDM; DFICR)
Loudspeaker
Loudspeaker Control bits (ARCR)
Loudspeaker output – pin
Loudspeaker output + pin
MCLK
MCLKR
MI3
MIC
MIN1/2
MIP1/2
Master Clock pin (synchronized system clock)
Master Clock Rate (SDICR)
Microphone input
Microphone Control bits (ATCR)
Microphone inputs – pins
Microphone inputs + pins
NOP
NOT
No Operation (CMDR)
No Test mode (TFCR)
PABX
PCI
PCM
PM
POR
PU
Private Automatic Branch Exchange
Peripheral Control Interface
Pulse Code Modulation
Piezo Mode; output to digital pins PZ1/PZ2 (TGSR)
Power-On Reset
Power-Up bit (GCR)
RAAR
RCM
RS
R/W
RX
Read Automatic Attenuation Receive
Reverse Channel Mode (CMDR)
Reset pin
Read/Write operation bit (CMDR)
Receive path
SA-SD
SCAE
SCI
SCLE
SCLK
SDI
SDR
SDX
SLOT
SM
SOP
SP
PCI I/O pins; I/O-control bits (SDICR)
Speech Comparator at the Acoustic Side
Serial Control Interface
Speech Comparator at the Line Side
Serial Clock pin (serial control interface)
Serial Data Interface
Serial Data Receive pin (serial control interface)
Serial Data Transmit pin (serial control interface)
IOM-2 Slot select for TE mode (GCR)
Stop Mode bit (TGCR)
Status Operation (CMDR)
Speakerphone enable bit (GCR)
Semiconductor Group
11
General Information
Table of Symbols (cont’d)
SQTR
S/T
Square/Trapezoid mode bit (TGCR)
Square/Trapezoid Generator
TE
TG
Terminal Equipment
Tone Generator bit (TGCR)
TR
TRL
TRR
TRX
TS
TX
Three party conferencing (VDM; DFICR)
Tone Ringing via Loudspeaker (TGSR)
Tone Ringing Receive bit (TGSR)
Tone Ringing Transmit bit (TGSR)
Time-Slot Selection in SDI-mode (TSCR)
Transmit path
VDD
VDDP
VREF
VSSA
VSSD
VSSP
Voltage supply (+ 5 V)
Analog Voltage supply for Power amplifiers (+ 5 V)
Voice Data Manipulation bits (DFICR)
Reference Voltage output pin
Analog ground (0 V)
Digital ground (0 V)
Analog ground for Power amplifiers (0 V)
WDF
Wave Digital Filter
XOP
Extended Operation (CMDR)
VDM
Semiconductor Group
12
Audio Ringing Codec Filter
Featuring Speakerphone Function
(ARCOFI®-SP)
PSB 2163
Preliminary Data
BICMOS-IC
1
Features
● Applications in digital terminal equipment featuring voice
functions
● Digital signal processing performs all CODEC functions
● Fully compatible to the G. 714 CCITT and ETSI (NET33)
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
specification
PCM A-Law/µ-Law (G. 711 CCITT) and 16-bit linear data
Flexible configuration of all internal functions
IOM-2 interface (TE- and non-TE-mode), Serial Control
Interface (SCI) and Serial Data Interface (SDI)
Three analog inputs for the microphone in the handset, the
speakerphone and the headset
Two differential outputs for a handset earpiece (200 Ω) and a
loudspeaker (50 Ω)
100-mW sine wave and 200-mW square wave loudspeaker
driver capability
Separate digital output for a piezo ringer
Flexible Peripheral Control Interface (PCI) in IOM-2 TE-mode
Flexible test and maintenance loopbacks in the analog front
end and the digital signal processor
Independent gain programmable amplifiers for all analog inputs
and outputs
Full digital speakerphone and monitoring support without any
external components (speakerphone test and optimization
function is available)
Two transducer correction filters
Side tone gain adjustment
Flexible DTMF, tone and ringing generator
Single 5-V power supply
Low power consumption: standby 1 mW, operating consumption is dependent on the selected operating mode
Advanced 1-µ BICMOS technology
P-DSO-28
P-LCC-28-R
P-DIP-28
Type
Ordering Code
Package
PSB 2163-T
Q67100-H6458
P-DSO-28 (SMD)
PSB 2163-N
Q67100-H6348
P-LCC-28-R (SMD)
PSB 2163-P
Q67100-H6460
P-DIP-28
Semiconductor Group
11
06.95
Features
Pin Configurations
(top view)
P-DSO-28
P-DIP-28
P-LCC-28-R
Semiconductor Group
14
Features
1.1
Pin Definitions and Functions
Function
Pin No. Symbol Input (I)
Output (O)
P-DSO
Open Drain (OD)
P-LCC
P-DIP
—
Power supply (5 V ± 5 %)
—
Power supply (5 V ± 5 %)
—
Digital Ground (0 V)
—
Analog Ground (0 V)
15
VDD
VDDP
VSSD
VSSA
VSSP
—
Analog Ground (0 V)
23
MODE
I
Mode Selection:
IOM-2 or serial control/data interface
25
AD
I
MCLK
I
IOM Address:
Chip address in IOM-2 two chip mode
Master Clock:
Synchronous system clock when serial
control/data interface is selected
24
RS
I
Reset:
A high signal on this pin forces the ARCOFI into
reset state
26
FSC
I
Frame Sync:
8-kHz frame synchronization signal (IOM-2 and
SDI-mode)
22
DCL
I
DCLK
I
DCL-System Clock:
1.536 MHz supplied by the application system
clock when IOM-2 mode is selected
DCLK Data Clock:
Data clock of the serial data interface (SDI)
DD
I/(OD)
DR
I
PZ1
PZ2
O
O
21
13
1
20
6
28
27
1)
1)
Data Downstream:
Receive data from layer-1 IOM-2 controlling
device
Data Receive:
Receive data of the serial data interface (SDI)
Digital Piezo Ringer Output:
When selected the tone ringer is routed to this
output (PZ1 & PZ2 are in opposite phases)
see DD/DU-voice channel swapping (XOP_D)
Semiconductor Group
15
Features
1.1
Pin Definitions and Functions (cont’d)
Function
Pin No. Symbol Input (I)
Output (O)
P-DSO
Open Drain (OD)
P-LCC
P-DIP
7
DU
OD/I
DX
OD/O
SD
IO
CS
I
SC
IO
SCLK
I
SB
IO
SDR
I
SA
IO
SDX
OD/O
19
VREF
O
2.4 V Output:
Output for biasing analog single ended inputs
8
9
MIP1
MIN1
I
I
Microphone Input 1:
This highly symmetrical differential input has
been designed for commonly used telephone
microphones
2
3
4
5
1)
2)
3)
1)
2)
Data Upstream:
Transmit data to the layer-1 IOM-2 controlling
device
Data Transmit:
Transmit data of the serial data interface (SDI)
Programmable I/O PCI Pin SD:
This port pin is only available in IOM-2 TE-mode
Chip Select:
A low level indicates a microprocessor access
to the ARCOFI-serial control interface (SCI)
Programmable I/O PCI Pin SC:
This port pin is only available in IOM-2 TE-mode
Serial Clock:
Clock signal of the serial control interface (SCI)
Programmable I/O PCI Pin SB:
This port pin is only available in IOM-2 TE-mode
Serial Data Receive:
Receive data line of the serial control interface
(SCI)
3)
Programmable I/O PCI Pin SA:
This port pin is only available in IOM-2 TE-mode
Serial Data Transmit:
Transmit data line of the serial control interface
(SCI)
see DD/DU-voice channel swapping (XOP_D)
programmable via bit SDICR.EPP0
programmable via bit SDICR.EPP1
Semiconductor Group
16
Features
1.1
Pin Definitions and Functions (cont’d)
Function
Pin No. Symbol Input (I)
Output (O)
P-DSO
Open Drain (OD)
P-LCC
P-DIP
11
10
MIP2
MIN2
I
I
Microphone Input 2:
This highly symmetrical differential input has
been designed for commonly used telephone
microphones
12
MI3
I
Microphone Input 3:
This single-ended input has been designed for
commonly used telephone microphones
14
16
LSP
LSN
O
O
Loudspeaker Output:
LSP & LSN are differential output pins which
can drive a 50-Ω loudspeaker directly; a piezo
transducer can also be used for ringing signal
instead of the loudspeaker
17
18
HOP
HON
O
O
Handset Earpiece Output:
HOP & HON are differential output pins which
can drive handset earpiece transducers directly
Semiconductor Group
17
Features
1.2
Logic Symbol
Figure 1
Logic Symbol of the ARCOFI®
Semiconductor Group
18
Features
1.3
Functional Block Diagram
Figure 2
Block Diagram of the ARCOFI®
Semiconductor Group
19
Features
1.4
System Integration
The complete family of ICs for digital terminals offered by Siemens simplifies the
development of these devices and gives a cost-effective solution to the design engineer.
The architecture of these terminals is based on a modular interface especially conceived
for ISDN and named IOM-2.
Figure 3 shows an example of an integrated multifunctional ISDN-S terminal using the
ISAC®-S TE. The ISAC-S TE (ISAC-S: ISDN S-Access controller PSB 2186) provides
the S interface and separates the B and D channels.
In this example one ICC (ICC: ISDN Communication Controller PEB 2070) is used to
handle data packets on the D-channel. A voice processor is connected to a
programmable digital signal processing codec filter (ARCOFI) via IC1 and a data
encryption module to a data device via IC2. B1 is used for voice communication and B2
for data communication.
Typical terminal applications are described in the next sections.
Figure 3
Example of ISDN-S Voice/Data Terminal
Semiconductor Group
20
Features
1.4.1
ISDN-Voice Terminal
Figure 4 shows a typical solution for a voice terminal for S interface.
The ARCOFI offers the functions of CODEC, filtering and speakerphone. It also carries
out the functions of tone ringing, DTMF, and A/D- and D/A-conversions. The ARCOFI
permits the direct connection of a handset and a speakerphone/loudspeaker.
The ARCOFI can be programmed and read out by the µC via the lOM-2-interface and
the ISAC-S TE. The same µC supervises the keyboard functions and the function
hook-on/off.
The S-interface functions such as activation/deactivation, clock recovery, clock
resynchronization as well as the layer-2 functions like LAPD-protocol handling are
executed by the ISDN-Subscriber Access Controller, also called ISAC-S TE PSB 2186.
A UK0-interface telephone can easily be derived from the voice terminal shown on
figure 4 by replacing the ISAC-S with the ISDN-Communication Controller ICC
PEB 2070 and the ISDN-Echo Cancellation Unit IEC PEB 2091.
A UP0-interface telephone is obtained by interchanging the IEC with the ISDN-Burst
Controller IBC PEB 2095. In this configuration the ICC PEB 2070 and the IBC PEB 2095
can be replaced by the ISAC-P TE PSB 2196.
Figure 5 shows a typical solution for a voice terminal for UK0- or UP0-interface.
In any case the whole terminal is power supplied either by the ISDN-Remote Power
Controller IRPC PSB 2120 or by the General Purpose Power Controller GPPC
PSB 2121.
Semiconductor Group
21
Features
Figure 4
Basic ISDN S-Voice Terminal
Semiconductor Group
22
Features
* ICC and IBC can be replaced by the ISAC®-P TE PSB 2196
Figure 5
Basic ISDN U-Voice Terminal
Semiconductor Group
23
Features
1.4.2
Terminal Adapter a, b for Analog Telephones
Figure 6 shows how to implement a terminal adapter (a, b) connecting analog
telephones to the ISDN-world. A SLIC can be connected to the ARCOFI.
The tip and ring information is transmitted transparently through the ARCOFI via the
C/I-channel of the lOM-channel 1, through the ISAC-S TE to the µC.
Figure 6
Terminal Adapter a, b for Analog Telephones
1.4.3
Voice/Data Terminal (PC-Card)
Figure 7 shows a voice/data terminal developed on a PC-card. The ITAC PSB 2110
(ITAC: ISDN-Terminal Adapter Circuit) ensures the bit rate adaptation necessary to
connect a non ISDN-terminal (V.24) to the ISDN-world.
The COM-IC is an UART (type: 8250 or 16450) which is necessary for modem
applications.
Semiconductor Group
24
Features
The Dual Port RAM is used for data transfer between the terminal processor and the PC.
The card is powered by the PC, and thus no power controller is necessary.
Figure 7
PC-Card as an ISDN-Voice/Data Terminal
1.4.4
Multifunctional ISDN-Terminal
Figure 8 gives an example of a multifunctional terminal. The HSCX SAB 82525 (HSCX:
High-Level Serial Communications Controller Extended) simplifies the realization of an
intelligent X.25 terminal adapter module whereas the ITAC PSB 2110 offers X.21, V.24,
V.110 or V.120 interfaces for non ISDN-terminals.
The µC connected to the ISAC-S TE PSB 2186 is the system master. The two other µCs
are the slaves. When a slave µC wants to intervene, it informs the master via the
C/l-channel of IOM-2 channel 1.
Semiconductor Group
25
Features
Figure 8
Multifunctional ISDN-Terminal
Semiconductor Group
26
Features
1.4.5
Digital Voice Terminal
The Serial Control Interface allows the ARCOFI to be programmed directly from a serial
port of a microcontroller.
The voice data may be transmitted via the IOM-2 interface or on a PCM-interface
provided from other transceiver devices. If the Serial Data Interface (SDI) is selected the
PCM-data rate can vary from 64 kbit/s up to 4096 kbit/s.
Figure 9 shows a PABX-voice terminal using the ISAC-P TE PSB 2196 together with a
Motorola type microcontroller. Figure 10 shows a PABX-voice terminal using a
transceiver device without IOM-2 interface.
Figure 9
UP0 PABX-Voice Terminal
Semiconductor Group
27
Features
Figure 10
PABX-Voice Terminal in Non-IOM®-2 Architecture
Semiconductor Group
28
Features
1.4.6
IOM®-2 Line Card Application
Some applications require the ARCOFI to connect directly to the IOM-2 interface of a line
card. The lOM-channel is selected via pin-strapping. The ARCOFI is programmed via
the MONITOR channel of the selected lOM-channel. Up to two ARCOFls can be
distinguished via AD input on the same IOM-channel.
This configuration allows control of up to 16 ARCOFls on one IOM-2 interface of a line
card controller.
Figure 11
ARCOFI® Line-Card Application
Semiconductor Group
29
Features
1.4.7
Primary Rate Application
The ARCOFI is designed to be connected to a 24 or 32 time-slot PCM-interface used
e.g. on primary rate equipment. The PCM-data is transmitted via the serial data interface
while programming of the ARCOFI is done on the serial control interface.
Figure 12
Primary Rate Application
Semiconductor Group
30
Features
1.4.8
Group 3 Fax / Modem Adapter
The ARCOFI can be connected to a standard fax or modem chip set designed for analog
networks. The ARCOFI converts the analog signal to PCM-data which are transmitted
over the digital network.
Figure 13
Group 3 Fax/Modem Adapter
Semiconductor Group
31
Functional Description
2
Functional Description
The ARCOFI bridges the gap between the audio world of microphones, earphones,
loudspeakers and the PCM-digital world by providing a full PCM-CODEC with all the
necessary transmit and receive filters. A block diagram of the ARCOFI is shown in
figure 14.
The ARCOFI can be subdivided in three main blocks:
● The ARCOFI Analog Front End (AFE)
● The ARCOFI Signal Processor (ASP)
● The ARCOFI Digital Interface (ADI)
A detailed description can be found in the following chapters.
Figure 14
Architecture of the ARCOFI®
Semiconductor Group
30
Functional Description
2.1
Analog Front End (AFE) Description
The Analog Front End section of the ARCOFI is the interface between the analog
transducers and the digital signal processor. In the transmit direction, the AFE-function
is to amplify the transducer input signals (microphones) and to convert them into digital
signals. In the AFE-receive section, the incoming digital signal is converted to an analog
signal which is output to an earpiece and/or a loudspeaker.
A block diagram of the AFE is shown in figure 15.
Figure 15
Signal Flow Graph of the AFE
Semiconductor Group
31
Functional Description
2.1.1
Description of the Analog I/O
Two differential inputs (MIP1/MIN1 and MIP2/MIN2) and one single-ended input (Ml3)
are connected to the amplifier AMI via an analog input multiplexer. The programmable
amplifier AMI provides a coarse gain adjustment range. Fine gain adjustment is
performed in the digital domain via the programmable gain adjustment stage GX (see
signal processor section). This allows a perfect level adaptation to various types of
microphone transducers without loss in the signal to noise performance.
Fully differential output HOP/HON connects the amplifier AHO to a handset earpiece.
Differential output LSP/LSN is provided for use with a 50-Ω loudspeaker. Up to 100 mW
(sine wave) of power can be delivered to the loudspeaker via the amplifier ALS. The
programmable amplifiers AHO and ALS provide a coarse gain adjustment range. Fine
gain adjustment is performed in the digital domain via the programmable adjustment
stage GR.
Two implemented AFE-configuration registers (ATCR & ARCR) provide a high flexibility
to accommodate an extensive set of user procedures and terminal attributes.
2.1.2
AFE-Attenuation Plan
Transmit Direction
Limit Values
Parameter
Transmit
0dBm0
max.
MIP1/MIN1
MIP2/MIN2
Microphone input level at max gain
AMI = 42 dB
1.33E-02
9.38E-03
– 42
– 38.33
MIP1/MIN1
MIP2/MIN2
Microphone input level at min gain
AMI = 0 dB
Unit
Reference
1.91E-02
1.35E-02
– 38.86
– 35.19
Vp
Vrms
dBm0
dBm
V
V
1.2 V
0.775 V
1.67E-00
1.18E-00
0
3.67
2.40E-00
1.70E-00
3.14
6.81
Vp
Vrms
dBm0
dBm
V
V
1.2 V
0.775 V
MI3
Input level at max gain
AMI = 24 dB
1.06E-01
7.46E-02
– 24
– 20.37
1.51E-01
1.07E-01
– 20.86
– 17.19
Vp
Vrms
dBm0
dBm
V
V
1.2 V
0.775 V
MI3
Input level at min gain
AMI = 0 dB
8.36E-01
5.91E-01
–6
– 2.33
1.20E-00
8.49E-01
– 2.86
0.81
Vp
Vrms
dBm0
dBm
V
V
1.2 V
0.775 V
Semiconductor Group
32
Functional Description
Receive Direction
Limit Values
Parameter
Receive
0dBm0
max.
LSP/LSN
Output level symmetrical
in a 50-Ω load
ALS = 2.5 dB
2.23E-00
1.58E-00
2.5
6.17
LSP/LSN
Output level symmetrical
in a 50-Ω load
ALS = – 21.5 dB
Unit
Reference
3.20E-00
2.26E-00
5.64
9.31
Vp
Vrms
dBm0
dBm
V
V
1.2 V
0.775 V
1.41E-01
9.95E-02
– 21.5
– 17.83
2.02E-01
1.43E-01
– 18.36
– 14.69
Vp
Vrms
dBm0
dBm
V
V
1.2 V
0.775 V
HOP/HON
Output level symmetrical
in a 200-Ω load
AHO = 2.5 dB
2.23E-00
1.58E-00
2.5
6.17
3.20E-00
2.26E-00
5.64
9.31
Vp
Vrms
dBm0
dBm
V
V
1.2 V
0.775 V
HOP/HON
Output level symmetrical
in a 200-Ω load
AHO = – 15.5 dB
2.81E-01
1.99E-01
– 15.5
– 11.83
4.03E-01
2.85E-01
– 12.36
– 8.69
Vp
Vrms
dBm0
dBm
V
V
1.2 V
0.775 V
Semiconductor Group
33
Functional Description
2.1.3
Interface to Acoustic Transducers
Note: ESD and EMV requirements are not included.
Figure 16
Example to Connect the AFE to Acoustic Transducers
Semiconductor Group
34
Functional Description
2.2
ARCOFI® Signal Processor (ASP) Description
The ARCOFI signal processor (ASP) has been conceived to perform all CCITT and ETSI
(NET33) recommended filtering in transmit and receive paths and is therefore fully
compatible to the G.714 CCITT and ETSI (NET33) specifications. The data processed
by the ASP is provided in the transmit direction by an oversampling A/D-converter
situated in the analog front end (AFE). Once processed, the speech signal is converted
into an 8 bit A-law or µ-law PCM-format or remains as a 16-bit linear word (2s
complement) if the compander is by-passed. The by-passing of the companding
depends on the bit setting in the configuration register DFICR (VDM-bits).
In the receive direction, the incoming PCM-stream is expanded into a linear format (if the
linear mode is selected, the expansion logic is by-passed) and subsequently processed
until it is passed to the oversampling D/A-converter.
Additionally to these standard codec functions, the ARCOFI provides a universal tone
generation unit and a high quality speakerphone function.
Semiconductor Group
35
Functional Description
Figure 17
Processor Signal Flow Graph
Semiconductor Group
36
Functional Description
2.2.1
Transmit Signal Processing
In the transmit direction a series of decimation filters reduces the sampling rate down to
the 8-kHz PCM-rate. These filters attenuate the out-of-band noise by limiting the transmit
signal to the voice band.
The decimation stages end with a EWDF low-pass filter which band-limits the voice
signal to the CCITT G.714 and ETSI (NET33) recommendations. The ARCOFI meets or
exceeds all the CCITT and ETSI (NET33) recommendations on attenuation distortion
and group delay distortion.
If the tone generation unit is connected to the transmit direction (TGSR.DTMF = 1), a
special 2-kHz DTMF-low-pass filter is placed in the transmit path. This filter guarantees
an attenuation of all unwanted frequency components, if DTMF-signals are transmitted.
Additionally, it is possible to add a programmable tone signal to the transmit voice signal
(TGSR.TRX = 1).
The GX-gain adjustment stage is digitally programmable allowing the gain to be
programmed from + 6 to 0 dB in steps of ≤ 0.25 dB (– ∞ dB and others are also possible).
Two bytes are necessary to set GX to the desired value. On reset, the GX-gain stage is
by-passed.
The transmit path contains a programmable high performance frequency response
correction filter FX allowing an optimum adaptation to different types of microphones
(dynamic, piezoelectric or electret). Twelve bytes are necessary to set FX to the desired
frequency correction function. On reset, the FX-frequency correction filter is by-passed.
Figure 18 shows the architecture of the FX/FR-filter.
A high-pass filter (HPX) is also provided to remove power line frequencies.
The voice signal, after being linearly processed, can be output as an 8-bit PCM-word
according to the CCITT G.711 A-law or the North-American µ-law format. If desired the
companding stage can be by-passed, a 16-bit linear word (2s complement) is then
output to the IOM-2 or SDI-interface.
2.2.2
Receive Signal Processing
In the receive path the incoming PCM-signal is expanded into a linear code according to
the selected A-law or µ-law. If the linear mode is chosen, the PCM-expander circuit is
by-passed and a 16-bit linear word (2s complement) has to be provided to the processor.
The block VDM offers several possibilities of voice/data manipulation for special
applications.
A programmable sidetone gain stage GZ adds a sidetone signal to the incoming voice
signal. The sidetone gain can be programmed from – 54 to 0 dB within a ± 1 dB tolerance
Semiconductor Group
37
Functional Description
range (– ∞ dB and others are also possible). Respectively two bytes are coded in the
CRAM to set GZ to the desired value. On reset, the GZ-gain stage is disabled (– ∞ dB).
A high-pass filter (HPR) is also provided to remove disturbances from 0 to 50/60 Hz due
to the telecommunication network.
The FR-frequency correction response filter is similar to the FX-filter allowing an
optimum adaptation to different types of loudspeakers or earpieces. Twelve bytes are
necessary to set FR to the desired frequency correction function. On reset, the
FR-frequency correction filter is by-passed.
The GR-gain adjustment stage is digitally programmable from – 6 to 0 dB in steps
≤ 0.25 dB (– ∞ dB and others are also possible). Respectively two bytes are coded in the
CRAM to set GR to the desired value. On reset, the GR-gain stage is by-passed.
A low-pass EWDF-filter limits the signal bandwidth in the receive direction according to
CCITT and ETSI (NET33) recommendations.
A series of low-pass interpolation filters increases the sampling frequency up to the
desired value.The last interpolator feeds the D/A-converter.
Figure 18
Architecture of the FX- and FR-Correction Filter
Semiconductor Group
38
Functional Description
2.2.3
Programmable Coefficients
This section gives a short overview of important programmable coefficients. For more
detailed information and about special applications, a special coefficient software
package is available (ARCOS-SP PLUS SIPO 2163).
Description of the programmable level adjustment parameters:
Parameter
# of CRAM
Bytes
Range
Comment
GX
2
12 to – ∞ dB Transmit gain adjustment
6 to 0 dB
Transmission characteristics guaranteed
GR
2
12 to – ∞ dB Receive gain adjustment
0 to – 6 dB Transmission characteristics guaranteed
GZ
2
12 to – ∞ dB Sidetone gain adjustment
Coefficients for GX, GR and GZ:
Gain [dB]
MSB
LSB
Gain [dB] MSB
LSB
Gain [dB] MSB
LSB
12.0
11.0
10.0
9.0
8.0
7.0
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
10H
10H
10H
01H
20H
30H
13H
B0H
A0H
23H
22H
23H
32H
B1H
B1H
33H
B2H
B3H
01H
31H
13H
4BH
94H
94H
51H
39H
49H
01H
B4H
12H
A4H
BCH
03H
39H
5AH
49H
0
– 0.5
– 1.0
– 1.5
– 2.0
– 2.5
– 3.0
– 3.5
– 4.0
– 4.5
– 5.0
– 5.5
– 6.0
– 7.0
– 8.0
– 9.0
– 10.0
– 11.0
A0H
B3H
A3H
A2H
BBH
BBH
BAH
BAH
A2H
AAH
9BH
AAH
AAH
B9H
9AH
9BH
9BH
93H
01H
42H
2BH
32H
4AH
13H
29H
5BH
01H
1BH
3AH
33H
22H
2CH
BCH
13H
32H
02H
– 12.0
– 13.0
– 14.0
– 15.0
– 16.0
– 17.0
– 18.0
– 19.0
– 20.0
– 21.0
– 22.0
– 23.0
– 24.0
– 25.0
– 26.0
–∞
A9H
9CH
99H
8CH
82H
84H
89H
8BH
84H
8CH
82H
84H
89H
8BH
84H
88H
01H
51H
13H
1BH
7BH
4BH
6AH
0CH
1CH
1CH
7CH
4CH
6BH
0DH
1DH
01H
Semiconductor Group
39
Functional Description
2.2.4
Tone Generation
2.2.4.1 Tone Generation Architecture
The ASP contains a universal tone generator which can be used for tone alerting, call
progress tones, DTMF-signals or other audible feedback tones.
For the receive channel, a universal switching to each signal path (earpiece,
loudspeaker and piezo ringer) is implemented. In the earpiece and loudspeaker
direction, an addition of the programmed tone sequence (sine-wave, trapezoid,
square-wave and DTMF) with the incoming voice signal is possible.
For the transmit direction, a supplementary DTMF-generator is implemented. If the
DTMF-generator is active (TGSR.DTMF = 1), only a part of the tone generator (TG) is
available for the receive direction (one or two tone sequences). In addition, a universal
switching to the transmit path is also possible (TGSR.TRX).
All the tone generation configurations are programmable in the registers TGCR and
TGSR (see description in chapter 4). A signal flow graph of the ARCOFI-tone generation
unit is shown in figure 19.
The tone generation can be subdivided into five main blocks:
●
●
●
●
●
Control Generator (CG)
Tone Generator (TG)
Tone Filter (TF)
Tone Level Adjustment (TLA)
DTMF-Generator (DTG)
A detailed description of the five main tone generation blocks follows in the next
sub-sections.
Semiconductor Group
40
Functional Description
Note: Adjustments in brackets are only available if the DTMF-generator is switched off
(TGCR.DTMF = 0).
Figure 19
Signal Flow Graph of the Tone Generation Unit
Semiconductor Group
41
Functional Description
2.2.4.2 Control Generator
In conjunction with the control generator it is possible to generate very complex signal
sequences without reprogramming the necessary parameters (e.g. pulsed three tone
calls). Four typical applications for the control generator programming are shown in
figure 20.
Figure 20
Typical Control Generator Application
Semiconductor Group
42
Functional Description
Function table of CG/TG-bit setting in TGCR:
TON/TOFF
CG
TG
Generator Output
X
X
TOFF
TON
0
0
1
1
0
1
X
X
No tone
Ringing sequence F1, F2, F3 without break
Break between two ringing sequences of F1, F2, F3
Ringing sequence until next break
Description of the programmable parameters:
Parameter
# of CRAM
Bytes
Range
Comment
TON
2
20 ms to 16 min
TOFF
2
20 ms to 16 min
Period while the tone generator
is turned on
Period while the tone generator
is turned off
2.2.4.3 Tone Generator
The tone generator contains a beat generator, a Square/Trapezoid generator, a second
trapezoid generator and an automatic stop for two and three tone ringing signals. With
the automatic stop function (SM-bit setting in TGCR) the multitone generation can be
stopped after a defined frequency. This avoids unpleasant sounds when stopping the
tone generator.
If the control generator is activated (TGCR.CG = 1) the bit setting of TG is insignificant.
Otherwise (TGCR.CG = 0) the TG-bit setting controls the activities of the tone generator.
A functional diagram of the tone generator is shown in figure 21.
Semiconductor Group
43
Functional Description
Figure 21
Functional Diagram of the Tone Generator
Distinctive alerting signals, allowing for example the use of different multitone ringing
patterns, are all programmable using the beat tone generator in conjunction with the
square/trapezoid generator. In the case of two or three tone ringing signals, the square/
trapezoid generator controls the output frequency pitch whilst the beat generator
controls the repetition rate. Either square or trapezoid shaped tones can be generated
depending on the TGCR.SQTR bit setting. If the piezo mode (PM or TRL in TGSR) is
chosen, only a square-wave is available (fixed amplitude of VDD). In this case the SQTR
bit in TGCR has no effect.
A secondary trapezoid generator is also built into the ARCOFI. Depending on the DT-bit
setting in the TGCR, the output signal of this generator is added to the output signal of
the Square/Trapezoid (S/T) generator. In conjunction with the S/T generator, a wide
variety of different dual tone signals can be programmed.
If the beat generator (TGCR.BT = 1) is enabled, the automatic stop function (SM-bit
setting in TGCR) can be activated. This prevents an uncontrolled turn-off of the tone
generator. Only when the generation of the frequency F2 or F3 (depending on the BM-bit
setting in TGCR) has been completed, the tone generator will switch off.
Semiconductor Group
44
Functional Description
Beat generator programming:
BT
BM
DT
Generator Output
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
Continuous signal
Continuous signal
Continuous signal
Continuous signal
Alternating signal
Alternating signal
Alternating signal
Alternating signal
F1, G1
F1, G1 + FD, GD1
F2, G2
F2, G2 + FD, GD2
F1, G1, T1; F2, G2, T2
F1, G1, T1; F2, G2, T2 + FD, GD1, T1; FD, GD2, T2
F1, G1, T1; F2, G2, T2; F3, G3, T3
F1, G1, T1; F2, G2, T2; F3, G3, T3 +
FD, GD1, T1; FD, GD2, T2; FD, GD3, T3
Description of the programmable parameters:
Parameter
# of CRAM
Bytes
Range
Comment
Fn
2/2/2
50 Hz to 4 kHz
Trapezoid shaped tone
16 kHz/m; (m ≥ 3)
Square-wave signal
Gn
1/1/1
0 dB to – 48 dB
Gain adjustment for
square/trapezoid generator
Tn
2/2/2
10 ms to 8 s
Period of time for two or three
tone sequences
FD
2
50 Hz to 4 kHz
Trapezoid shaped tone
GDn
1/1/1
0 dB to – 48 dB
Gain adjustment for
trapezoid generator
n is either 1, 2 or 3
Note: 0-dB gain setting of G1, G2 or G3 and GD1, GD2 or GD3 corresponds to the
maximum PCM-level (A-Law: + 3.14 dB)
Semiconductor Group
45
Functional Description
2.2.4.4 Tone Filter
The tone filter contains a programmable equalizer and a saturation amplifier (see
figure 19). If no filter function is necessary, a by-pass mode can be used
(TGCR.ETF = 0). A brief description of the tone filter follows below.
The equalizer is realized as a band-pass filter. The filter parameters (center frequency,
bandwidth, and attenuation of the stop-band) are programmable.
A generated square-wave or trapezoid signal can be converted by the equalizer into a
sine-wave signal. A maximum attenuation of the first harmonic frequency of 50 dB is
possible.
By programming the equalizer as a broadband filter, the quality of the DTMF-signal
(receive direction) is improved. A level balancing of the two frequency components can
be made with G1, G2, G3 and GD1, GD2, GD3.
The two main purposes of the programmable saturation amplification are:
● Level balancing of the filtered signal (avoidance of overload effects).
● Amplification up to + 12 dB followed by a saturation of the incoming signal. This
saturation amplification converts a sine-wave signal into a square-wave or a trapezoid
signal where their edges are eliminated. This method produces pleasant ringing
tones.
Description of the programmable parameters:
Parameter
# of CRAM
Bytes
Range
Comment
A1
A2
1
1
200 Hz to 4 kHz
0 to – 1
K
GE
1
1
0 to 54 dB
+ 12 to – 12 dB
Center frequency
Bandwidth
(strongly dependent on A1 and K)
Attenuation of the stop-band
Saturation amplification
Semiconductor Group
46
Functional Description
2.2.4.5 Tone Level Adjustment
The two level adjustment stages GTR and GTX determines the output levels of the tone
generation (see figure 19).
Description of the programmable parameters:
Parameter
# of CRAM
Bytes
Range
Comment
GTX
1
GTR
1
0 dB to – 50 dB
(also – ∞ dB)
0 dB to – 50 dB
(also – ∞ dB)
Level adjustment for the output which
is connected to the transmit channel
Level adjustment for the output which
is connected to the receive channel
2.2.4.6 DTMF-Generator (transmit)
The DTMF-generator contains two independent trapezoid generators which can be
programmed in a wide frequency and gain range. If the DTMF-generator is active
(TGSR.DTMF = 1), the output signal is automatically switched to the transmit direction.
In this case the attenuation of the unwanted frequency components is executed by a
special DTMF-low-pass filter to the following limits:
Frequency Band
Min. Attenuation
0 – 300 Hz
300 – 3400 Hz
3400 – 4000 Hz
33 dB
20 dB
33 dB
The pre-emphasis of 2 dB between the high and the low DTMF-frequency groups has to
be set with the independent gain stages for the two trapezoid generators (G3 and GD3).
All generated DTMF-frequencies are guaranteed within a ± 1 % deviation.
Semiconductor Group
47
Functional Description
DTMF-frequency (F3, FD) programming:
CCITT Q.23 ARCOFI® Nominal Relative Deviation
Low Group
697
770
852
941
High Group
1209
1336
1477
1633
Coefficients
from CCITT
high
low
697.1
770.3
852.2
941.4
+ 143 ppm
+ 390 ppm
+ 235 ppm
+ 425 ppm
4F
A6
45
20
16
18
1B
1E
1209.5
1336.9
1477.7
1632.8
+ 414 ppm
+ 674 ppm
+ 474 ppm
– 122 ppm
B4
C8
49
40
26
2A
2F
34
Note: The deviations due to the inaccuracy of the incoming clock DCL/MCLK, when
added to the nominal deviations tabulated above give the total absolute deviation
from the CCITT-recommended frequencies.
Description of the programmable parameters:
Parameter
# of CRAM
Bytes
Range
Comment
F3
G3
2
1
50 Hz to 4 kHz
0 dB to – 48 dB
FD
GD3
2
1
50 Hz to 4 kHz
0 dB to – 48 dB
Trapezoid shaped tone 1
Gain adjustment
for trapezoid generator 1
Trapezoid shaped tone 2
Gain adjustment
for trapezoid generator 2
2.2.5
ARCOFI® Speakerphone Support
The speakerphone option of the ARCOFI-SP PSB 2163 performs all voice switching
functions without any external components, just by software. All these operational
functions realized by the signal processor are completely parameterized. This technique
offers a high level of flexibility and reproducibility.
There are three modes of operation: “speech mode”, “listen mode”, and “idle mode”. In
the speech mode the receive path is attenuated while in listen mode the attenuation is
switched to the transmit path. In the idle mode the attenuation is halved between
transmit and receive paths. The switching is mainly controlled by the speech
comparators while speech activity is recognized by the speech detectors.
Semiconductor Group
48
Functional Description
As the signal flow graph of the speakerphone option shows (figure 22), the complete
operational algorithm is situated between the Analog Front End/Signal Processing and
the compression/expansion logic. This has the advantage that the speakerphone
function is independent of any country specific transmission characteristics. Thus
telephone sets can be optimized and adjusted to the particular geometrical and acoustic
environment.
The main features of the speakerphone signal processing are:
● Two separate attenuation stages activated by voice, one for the transmit and one for
the receive path. They are controlled by the current and past speech activities.
● Immediate mode switching mainly controlled by two comparators, one at the acoustic
side and one at the line side.
● Speech detection by special speech detectors in the respective transmit and receive
directions. Different time constants are separately programmable for signal and
noise.
● Background noise monitoring to eliminate continuous background noise from speech
control. All time constants are user programmable.
Figure 22
Speakerphone Signal Flow Graph of the ARCOFI®
Semiconductor Group
49
Functional Description
2.2.5.1 Speech Detector
The speech detectors (figure 23) contained in both transmit and receive directions
consist of two main blocks:
● Background Noise Monitor (BNM) and
● Signal Processing
Figure 23
Speech Detector Signal Flow Graph
Semiconductor Group
50
Functional Description
Background Noise Monitor
The tasks of the noise monitor are to differentiate voice signals from background noise,
even if it exceeds the voice level, and to recognize voice signals without any delay.
Therefore the Background Noise Monitor consists of the Low-Pass Filter 2 (LP2) and the
offset in two separate branches. Basically it works on the burst-characteristic of the
speech: voice signals consist of short peaks with high power (bursts). In contrast,
background noise can be regarded approximately stationary from its average power.
Low-Pass Filter 2 provides different time constants for noise (non-detected speech) and
speech. It determines the average of the noise reference level. In case of background
noise the level at the output of LP2 is approximately the level of the input. Due to the
offset OFF the comparator remains in the initial state. In case of speech at the
comparator input the difference between the signal levels of the offset branch and of the
LP2-branch increases and the comparator changes state. At speech bursts the digital
signals arriving at the comparator via the offset branch change faster than those via the
LP2-branch so that the comparator changes its polarity. Hence two logical levels are
generated: one for speech and one for noise.
A small fade constant (LP2N) enables fast settling down the LP2 to the average noise
level after the end of speech recognition. However, a too small time constant for LP2N
can cause rapid charging to such a high level that after recognizing speech the danger
of an unwanted switching back to noise exists. It is recommended to choose a large
rising constant (LP2S) so that speech itself charges the LP2 very slowly. Generally, it is
not recommended to choose an infinite LP2S because then approaching the noise level
is disabled. During continuous speech or tones the LP2 will be charged until the
limitation LP2L is reached. Then the value of LP2 is frozen until a break discharges the
LP2. This limitation LP2L of this charging especially on the RX-path permits transmission
of continuous tones and “music on hold”.
The offset stage represents the exact level threshold in [dB] between the speech signal
and averaged noise.
Signal Processing
As described in the preceding chapter, the Background Noise Monitor is able to
discriminate between speech and noise. In very short speech pauses e.g. between two
words, however, it changes immediately to non-speech, which is equal to noise.
Therefore a peak detection is required in front of the Noise Monitor.
The main task of the Peak Detector is to bridge the very short speech pauses during a
monolog so that this time constant has to be long. Furthermore, the speech bursts are
stored so that a sure speech detection is guaranteed. But if no speech is recognized the
noise low-pass LP2 must be charged rapidly to the average noise level. Additionally the
noise edges are to be smoothed. Therefore two time constants are necessary and are
Semiconductor Group
51
Functional Description
separately programmable: PDS for speech and PDN for space (background noise)
signals.
The Peak Detector is very sensitive to spikes. The LP1 filters the incoming signal
containing noise in a way that main spikes are eliminated. Due to the programmable time
constant it is possible to refuse high-energy sibilants and noise edges.
To compress the speech signals in their amplitudes and to ease the detection of speech,
the signals have to be companded logarithmically. Hereby, the speech detector should
not be influenced by the system noise which is always present but should discriminate
between speech and background noise. The limitation of the logarithmic amplifier can
be programmed via the parameter LIM, where the upper half-byte features LIMX and the
lower half-byte LIMR. LIM is related to the maximum PCM level. A signal exceeding the
limitation defined by LIM is getting amplified logarithmically, while very smooth system
noise below is neglected. It should be the level of the minimum system noise which is
always existing; in the transmit path the noise generated by the telephone circuitry itself
and in receive direction the level of the first bit which is stable without any speech signal
at the receive path.
Description of the programmable speech detector parameters:
Parameter
# of CRAM
Bytes
Range
Comment
LP1
OFF
PDS
PDN
LP2S
LP2N
LP2L
LIMX, LIMR
1
1
1
1
1
1
1
1
1 to 512 ms
0 to 50 dB
1 to 512 ms
1 to 512 ms
4 to 2000 s
1 to 512 ms
0 to 95 dB
– 36 to – 78 dB
Time constant LP1
Level offset up to detected noise
Time constant PD (signal)
Time constant PD (noise)
Time constant LP2 (signal)
Time constant LP2 (noise)
Limitation of LP2, related to LIM
Limitation of logarithmic amplifier
2.2.5.2 Speech Comparators (SC)
Switching from one active mode to another one is mainly controlled by the speech
comparators. There are two Speech Comparators, one at the acoustic (AE) and one at
the line side (LE). This offers a different programming of the sensitivity of the speech
detectors and avoids clipping due to echoes. These comparators continuously compare
the signal levels of both signal paths and control the effect of the echos at the acoustic
side and the line side. Once speech activity has been detected, the comparator switches
at once in that direction in which the speech signal is stronger. For this purpose each
signal is compared to the sum of the other and the returned echo.
Semiconductor Group
52
Functional Description
Speech Comparator at the Acoustic Side (SCAE)
In principle, the SCAE works according to the following equation:
if
SX > SR + VAE then TX
else RX
Being in RX-mode, the speech comparator at the acoustic side controls the switching to
TX-mode. Only if the SX-signal is higher than the SR-signal plus the expected/measured
acoustic level enhancement (VAE), the comparator switches immediately to TX-mode.
Physically the level enhancement (VAE) is divided into two parts: GAE and GDAE.
Figure 24
Speech Comparator at the Acoustic Side
Semiconductor Group
53
Functional Description
At the SCAE-input, logarithmic amplifiers compress the signal range. Hence after the
required signal processing for controlling the acoustic echo, pure logarithmic levels on
both paths are compared.
Principally, the main task of the comparator is to control the echo. The internal coupling
due to the direct sound and mechanical resonances are covered by GAE. The external
coupling, mainly caused by the acoustic feedback, is controlled by GDAE/PDAE.
The Gain of the Acoustic Echo (GAE) corresponds to the terminal couplings of the
complete telephone: GAE is the measured or calculated level enhancement between
both receive and transmit inputs of the SCAE (refer to figure 22). It equals the sum of
the amplification of ALS plus the gain due to the loudspeaker/microphone coupling plus
the TX-amplification of AMI and GX. To succeed in a sure differentiation between
original speech and echo, it must be guaranteed that the TX-signal does not run into
saturation due to the loudspeaker/microphone coupling. Therefore, it is recommended
to reduce the TX-gain by 10 dB in front of the SCAE at least in the loudest loudspeaker
volume step. To fulfill the sending loudness rating, this gain is realized by the LGAX/
AGCX which follows the SCAE. Of course, the GAE has to be reduced by the same
amount.
To control the acoustic feedback two parameters are necessary: GDAE-features the
actual reserve on the measured GAE. Together with the Peak Decrement (PDAE) it
simulates the echo behaviour at the acoustic side: After RX-speech has ended there is
a short time during which hard couplings through the mechanics and resonances and
the direct echo are present. Till the end of that time (∆t) the level enhancement VAE must
be at least equal to GAE to prevent clipping caused by these internal couplings. Then,
only the acoustic feedback is present. This coupling, however, is reduced by air
attenuation. For this in general the longer the delay, the smaller the echo being valid.
This echo behaviour is featured by the decrement PDAE.
Semiconductor Group
54
Functional Description
Figure 25
Interdependence of GDAE and PDAE
According to figure 25, a compromise between the reserve GDAE and the decrement
PDAE has to be made: a smaller reserve (GDAE) above the level enhancement GAE
requires a longer time to decrease (PDAE). It is easy to overshout the other side but the
intercommunication is harder because after the end of the speech, the level of the
estimated echo has to be exceeded. In contrary, with a higher reserve (GDAE*) it is
harder to overshout continuous speech or tones, but it enables a faster
intercommunication because of a stronger decrement (PDAE*).
Two pairs of coefficients, GDSAE/PDSAE when speech is detected, and GDNAE/
PDNAE in case of noise, offer a different echo handling for speech and non-speech.
With speech, even if very strong resonances are present, the performance will not be
worsened by the high GDSAE needed. Only when speech is detected, a high reserve
prevents clipping. A time period ETAE [ms] after speech end, the parameters of the
comparator are switched to the “noise” values. If both sets of the parameters are equal,
ETAE has no function.
Semiconductor Group
55
Functional Description
Description of the programmable parameters:
Parameter
# of CRAM
Bytes
Range
Comment
GAE
GDSAE
PDSAE
1
1
1
– 48 to + 48 dB
0 to 48 dB
0.16 to 42 ms/dB
GDNAE
PDNAE
1
1
0 to 48 dB
0.16 to 42 ms/dB
ETAE
1
0 to 1020 ms
Gain of Acoustic Echo
Reserve when speech is detected
Peak Decrement
when speech is detected
Reserve when noise is detected
Peak Decrement
when noise is detected
Echo time
Speech Comparator at the Line Side (SCLE)
Principally, the SCLE works similarly to the SCAE. The formula of SCLE is the following:
if
SR > SX + VLE then RX
else TX
Being in TX-mode, the speech comparator at the line side controls the switching to
RX-mode. When the SR-signal is higher than the SX-signal plus the expected/measured
echo return loss (VLE) and if SDR has detected speech, the comparator switches
immediately to RX-mode.
Semiconductor Group
56
Functional Description
Figure 26
Speech Comparator at the Line Side
The Gain of the Line Echo (GLE) directly corresponds to the echo return loss of the link.
Generally, it is specified to 27 dB. However, the worst case loss can be estimated to
10 dB. This means, the echo returns at least attenuated by 10 dB. The coefficient GLE
should be programmed with an extra reserve of 2 dB so that very smooth noise is
processed correctly.
Similarly to the acoustic side, GDLE at the line side features the reserve above GLE
which is necessary to control the echo via the decrement PDLE. GDLE and PDLE are
interdependent. Exactly ∆t [ms] after the end of RX-speech the level enhancement VLE
must be at least GLE to prevent clipping.
Semiconductor Group
57
Functional Description
Two pairs of coefficients are available: GDSLE/PDSLE while speech is detected and
GDNLE/PDNLE in case of noise. This offers the possibility to control separately the
far-end echo during speech and the near-end echo while noise is detected. However,
this requires an attenuation between the speech detectors SDX and SDR: If the SDX
does not recognize any speech, the SDR must not detect speech due to the far-end
echo. Note, that LIMX and LIMR are also influencing the sensitivity of the speech
detection. ETLE [ms] after the final speech detection the parameter sets are switched. If
both sets are equal, ETLE has no meaning.
Description of the programmable parameters:
Parameter
# of CRAM
Bytes
Range
Comment
GLE
GDSLE
PDSLE
1
1
1
– 48 to + 48 dB
0 to 48 dB
0.16 to 42 ms/dB
GDNLE
PDNLE
1
1
0 to 48 dB
0.16 to 42 ms/dB
ETLE
1
0 to 1020 ms
Gain of Line Echo
Reserve when speech is detected
Peak Decrement
when speech is detected
Reserve when noise is detected
Peak Decrement
when noise is detected
Echo time
2.2.5.3 Attenuation Control Unit
The Attenuation Control unit controls the attenuation stages GHX of the transmit and
GHR of the receive directions respectively. The programmable loss is switched either
completely to a single path or, in the “IDLE” mode, is halved to each direction.
In addition, attenuation is also influenced by the Automatic Gain Control stages (AGCX
and AGCR): For the total loop gain never to exceed 1, the sweep range (of ATT) is
automatically enlarged with high-gain amplification of the AGCs while it will be
accordingly reduced with low-gain.
Changing from one speakerphone mode into another one depends on the
determinations of one comparator plus the corresponding speech detector. Hence
attenuation is influenced by the current and past speech activities. Also rate of change
varies: changing from “speech mode” or “listen mode” to “idle mode” is programmable
by the rate factor DS. Direct changes from “speech mode” to “listen mode” or vice-versa
and changes from “idle mode” to “speech mode” or “listen mode” can be programmed
via the factor SW in a large range.
Semiconductor Group
58
Functional Description
Description of the programmable parameters:
Parameter
# of CRAM
Bytes
Range
Comment
TW
ATT
1
1
16 ms to 4 s
0 dB to 95 dB
DS
1
SW
1
Wait time
Attenuation programmed in GHR or
GHX if speech activity for the other
side was detected
Decay Speed
0.6 to 680 ms/dB
(Decay Time TD = DS × ATT/2)
0.0052 to 10 ms/dB Switching time (dependent on ATT)
2.2.5.4 Speakerphone Test Function
The ARCOFI offers a test mode to ease the optimization of the switching behaviour
(TFCR.EPZST = 1). This function can also be used for signalling e.g. the speech mode
during a normal telephone conversation. This mode uses the piezo pins PZ1 and PZ2.
The PZ1 pin forced to a high level indicates that neither of the speech detectors
recognizes speech (refer to TW, DS, and idle state); when any speech activity is
detected, this pin is at a low level. At the PZ2 pin a logical “1” indicates the speech mode
(TX-mode) while a “0” signals listen mode (RX mode).
2.2.5.5 Automatic Gain Control of the Transmit Direction (AGCX)
An AGCX is inserted into the transmit path (figure 27) to reach nearly constant loudness
ratings independent of the varying distance between the speaker and the microphone.
Regulation range is between 0 dB and + 12 dB which must be enabled by setting
TGSR.PM1 = 1 (Piezo-Mode).
Operation of the AGCX depends on a threshold level. Its value (via parameter COMX)
corresponds to a signal level relative to the maximum PCM-value at which the
microphone signal is to become amplified. Regulation follows two time constants:
TMHX, which limits the signal amplitude, is shorter because the signal has to be reduced
before it goes into saturation. TMLX, which amplifies the signal if it falls below the
reference level, is greater because this threshold effect should hardly be perceptible.
For reasons of physiological acceptance the AGCX is automatically reduced in case of
continuous background noise e.g. by ventilators. The reduction is programmed via the
NOlSX-parameter. When the noise level increases the threshold determined by NOISX,
the amplification AGX will be reduced by the same amount the noise level is above the
threshold.
Semiconductor Group
59
Functional Description
A programmable Loudness Gain Adjustment stage (LGAX) offers the possibility to
amplify the TX-signal after the speech comparator SCAE and the speech detector SDX.
If a lower signal range in front of the SDX is necessary to determine between speech
and echo a part of the TX-amplification can be transferred to the LGAX. It is enabled
together with the bit GCR.SP. Even if the AGCX is disabled in speakerphone mode the
LGAX remains enabled.
Figure 27
Function of the Transmit AGC
Description of the programmable parameters:
Parameter # of CRAM
Bytes
LGAX
COMX
AGX
TMLX
TMHX
NOISX
1
1
1
1
1
1
Semiconductor Group
Range
Comment
– 12 to 12 dB
0 to – 73 dB
0 to 18 dB
1 to 2700 ms/dB
1 to 340 ms/dB
0 to – 95 dB
Loudness Gain Adjustment
Compare level rel. to max. PCM-value
Gain range of Automatic control
Settling time constant for lower levels
Settling time constant for higher levels
Threshold for AGC-reduction
by background noise
60
Functional Description
2.2.5.6 Automatic Gain Control of the Receive Direction (AGCR)
The Automatic Gain Control of the receive direction AGCR (figure 28) is similar to the
transmit AGC. One additional parameter (AAR) offers an automatic amplification. The
maximum attenuation is selectable with AAR. Depending on the parameters AAR and
AGR three different behaviours of the AGCR are possible:
• AGR = 0:
the only task of the AGCR is to prevent clipping. If the RX-signal
exceeds the compare level, the AGCR starts to attenuate it.
• AGR = AAR:
the AGCR works as an automatic amplifier. The RX-signal will be
amplified if it is smaller than the compare level.
• AAR > AGR: the combination of the previous tasks: If the signal is smaller than
the compare level it will be amplified while if it exceeds the level it
will be attenuated. The AGCR functions like a dynamic compressor.
While the digital RX-signal level varies, the volume coming out of the
loudspeaker remains constant.
If the AGCR is disabled in speakerphone mode, the Loudness Gain Adjustment stage
(LGAR) offers a programmable amplification in front of the SDR. It is enabled together
with the bit GCR.SP. It is highly recommended to program reasonable amplifications in
the digital gain stages. Otherwise the ASP will run into saturation above the 3.14 dB
PCM-value.
Note that the speech detector for the receive direction is supplied with the signal that
comes out of the AGR-block.
Figure 28
Function of the Receive AGC
Semiconductor Group
61
Functional Description
Description of the programmable parameters:
Parameter # of CRAM
Bytes
LGAR
COM
AAR
AGR
TMLR
TMHR
NOISR
1
1
1
1
1
1
1
Range
Comment
– 12 to 12 dB
0 to – 73 dB
0 to – 47 dB
0 to 18 dB
1 to 2700 ms/dB
1 to 340 ms/dB
0 to – 95 dB
Loudspeaker Gain Adjustment
Compare level rel. to max. PCM-value
Attenuation range of Automatic control
Gain range of Automatic control
Settling time constant for lower levels
Settling time constant for higher levels
Threshold for AGC-reduction
by background noise
2.2.5.7 Loudhearing
The ARCOFI-SP offers the possibility to do a so called “controlled monitoring” when the
bit ARCR.CME is set. This mode can only be used together with the speakerphone
mode (GCR.SP) With CME = 1 the attenuation stage GHR is fixed to a value of 0 dB but
the attenuation takes place in the analog loudspeaker amplifier ALS in a way that the
amplification of the ALS is set to – 9.5 dB as soon as the attenuation control unit switches
to transmit mode. Therefore in transmit direction the same behaviour as in
speakerphone mode occurs but in the receive direction the handset output offers a
signal as in normal handset mode while the volume at the loudspeaker output will be
reduced to a low level during transmit mode. If the programming for the loudspeaker
output (ARCR.LSC) is already chosen for values of less or equal – 9.5 dB, no further
attenuation takes place.
Semiconductor Group
62
Functional Description
2.2.6
Speakerphone Coefficient Set
This example shows a possible configuration for a speakerphone application. All
described coefficients can be used as a basic programming set.
CMD Sequence
Coefficient
Code
Value
COP_A
COP_A
COP_A
COP_A
COP_A
COP_A
COP_A
COP_A
GAE
GLE
ATT
ETAE
ETLE
TW
DS
SW
0CH
E5H
40H
0CH
32H
09H
25H
64H
4.50 dB
– 10.02 dB
24.00 dB
48.00 ms
200.00 ms
144.00 ms
– 99 ms/dB
0.6 ms/dB
COP_B
COP_B
COP_B
COP_B
COP_B
COP_B
COP_B
COP_B
GDSAE
PDSAE
GDNAE
PDNAE
GDSLE
PDSLE
GDNLE
PDNLE
20H
06H
20H
06H
40H
02H
40H
02H
6.02 dB
7.1 ms/dB
6.02 dB
7.1 ms/dB
12.00 dB
21.3 ms/dB
12.00 dB
21.3 ms/dB
COP_C
COP_C
COP_C
COP_C
COP_C
COP_C
COP_C
COP_C
LIMX, LIMR
OFFX
OFFR
LP2LX
LP2LR
LP1X
LP1R
reserved 00H
22H
0CH
0CH
20H
20H
E1H
E1H
– 48.16 dB, – 48.16 dB
4.50 dB
4.50 dB
12 dB
12 dB
4.00 ms
4.00 ms
COP_D
COP_D
COP_D
COP_D
COP_D
COP_D
COP_D
COP_D
PDSX
PDNX
LP2SX
LP2NX
PDSR
PDNR
LP2SR
LP2NR
26H
F4H
20H
44H
26H
F4H
20H
44H
102.34 ms
32.00 ms
6.55 s
30.06 ms
102.34 ms
32.00 ms
6.55 s
30.00 ms
Semiconductor Group
63
Functional Description
Speakerphone Coefficient Set (cont’d)
CMD Sequence
Coefficient
Code
Value
COP_E
COP_E
COP_E
COP_E
COP_E
COP_E
COP_E
COP_E
LGAX
COMX
AGX
TMHX
TMLX
NOISX
reserved 00H
reserved 00H
01H
C3H
01H
0AH
24H
4FH
9.60 dB
– 20.43 dB
12.04 dB
14.00 ms/dB
383.00 ms/dB
– 66.23 dB
COP_F
COP_F
COP_F
COP_F
COP_F
COP_F
COP_F
COP_F
LGAR
COMR
AAR
AGR
TMHR
TMLR
NOISR
reserved 00H
F0H
B2H
55H
00H
0AH
2FH
4FH
6.04 dB
– 15.05 dB
– 33.16 dB
18.06 dB
13.95 ms/dB
500.84 ms/dB
– 66.23 dB
Semiconductor Group
64
Functional Description
2.3
ARCOFI® Digital Interface (ADI)
The ADI-function consists of two interface blocks:
● The Peripheral Control Interface (PCI) or the Serial Control Interface (SCI)
● The IOM-2 interface (TE- or non-TE-timing mode) or the Serial Data Interface (SDI)
Supplementary functions are accessed by strapping the pins MODE and AD according
to the following table:
Pin MODE
Pin AD
Mode
Description
0
0
0
1
1
1
0
1
MCLK
0
1
MCLK
IOM-2 TE
IOM-2 TE
Test
IOM-2 non-TE
IOM-2 non-TE
SDI
IOM-2 TE-timing mode (AD = 0)
IOM-2 TE-timing mode (AD = 1)
IOM-2 Non-TE-timing mode (AD = 0)
IOM-2 Non-TE-timing mode (AD = 1)
Serial Data Interface
A detailed description is in the following chapter.
2.3.1
PCI-Interface
The Peripheral Control Interface (PCI) provides 4 programmable l/O-pins to control the
peripheral devices (for more detailed information see section 4, DFICR). These four
interface pins are only available in the IOM-2 terminal mode (TE-mode).
Otherwise these pins are used as slot select pins in the IOM-2 non-TE-timing mode or
used as a Serial Control Interface (SCI).
SA-SD
Mode
PCI
Slot Select
SCI
IOM-2 TE
IOM-2 Non TE
Serial Mode
2.3.2
IOM®-2 Frame Structure and Timing Modes
This interface consists of one data line per direction (DD: Data Downstream; DU: Data
Upstream). Two additional signals define the data clock (DCL) and the frame
synchronization (FSC).
In terminal applications, the IOM-2 constitutes a powerful backplane bus offering
intercommunication and sophisticated control capabilities for peripheral modules (e.g.
ARCOFI).
Semiconductor Group
65
Functional Description
The channel structure of the IOM-2 is described in figure 29.
B1
B2
MONITOR
D
C/I
MM
R X
Figure 29
Channel Structure of IOM®-2
● The 64-kbit/s channels, B1 and B2, are conveyed in the first two bytes.
● The third byte (monitor channel) is used for programming and controlling devices
attached to the IOM-2 interface.
● The fourth byte (control channel) contains two bits for the 16-kbit/s D-channel, four
command/indication bits for controlling activation/deactivation and for additional
control functions, two bits MR and MX for supporting the handling of the MONITOR
channel.
In case of an IOM-2 interface the frame structure depends on whether TE- or
non-TE-mode is selected.
Non-TE-Timing Mode
The frame of this mode is a multiplex of eight IOM-2 channels (figure 30), each channel
has the structure as shown in figure 29.
The ARCOFI is assigned to one of eight channels (0 to 7) by strapping SB to SD
according to the following table:
Pin SD
Pin SC
Pin SB
Selected IOM®-2 Channel
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
0
1
2
3
4
5
6
7
Pin SA is not used in this mode and should be connected to VDD or VSSD.
Thus the data rate per channel is 256 kbit/s, whereas the bit rate is 2.048 kbit/s. The
IOM-2 interface signals are:
DD, DU : 2048 kbit/s
DCL
: 4096 kHz (double clock rate)
FSC
: 8 kHz
Semiconductor Group
66
Functional Description
Figure 30
Multiplexed Frame Structure of the IOM®-2 Interface in Non-Terminal Timing Mode
Semiconductor Group
67
Functional Description
TE-Timing Mode
The IOM-2 frame provides three complete IOM channels (figure 31):
● Channel 0 contains 144 kbit/s (2B + D) plus monitor and command/indication
channels for the layer-1 device.
● Channel 1 contains two 64-kbit/s intercommunication channels plus monitor and
command/indication channels for other IOM-2 devices (e.g. ARCOFI).
● Channel 2 is used for D-channel arbitration.
The IOM-2 signals are:
DD, DU : 768 kbit/s
DCL
: 1536 kHz (double clock rate)
FSC
: 8 kHz
B1, B2
MON1
C/I1
IC1, IC2
Bearer voice data channel 1/2 to/from layer-1 device
Monitor channel 1
Command/Indicate channel 1
Intercommunication channel 1/2
Figure 31
IOM®-2 Interface Structure in Terminal Mode
Semiconductor Group
68
Functional Description
2.3.3
Serial Control Interface
When the MODE pin is tied high and the AD/MCLK pin is used as system clock input
(MCLK), the internal configuration registers and the coefficient RAM of the ARCOFI are
programmable via the serial control interface. It consists of 4 lines: SCLK, SDR, SDX
(open drain or push-pull) and CS.
CS is used to start a serial access to the ARCOFI-registers and the coefficient RAM.
Following a falling edge on CS, the first eight bits transmitted on SDR specify the
command. The subsequent one, two, four or eight bytes (depending on command)
read(s) or write(s) the contents of the selected registers or RAM-locations until the CS
line becomes inactive. If a read command is chosen, the first byte after the command is
the identification code of the ARCOFI-SP PSB 2163 (see also chapter 3.4.2.1). After
one command sequence is completed at least one NOP-command is required (see
figure 32).
A transfer sequence can be broken by setting CS high. All bytes already sent when CS
changes to high are valid.
The data transfer is synchronized by the SCLK input. SDX changes with the falling edge
of SCLK while the contents of SDR is latched on the rising edge of SCLK.
Figure 32 shows the timing of a serial control interface transfer (one byte transfer).
Semiconductor Group
69
Functional Description
Figure 32
Serial Control Interface Timing
Semiconductor Group
70
Functional Description
2.3.4
Serial Data Interface
If the serial control interface is selected, the ARCOFI supports an additional serial data
interface for B-channel transfer. This control interface consists of five lines: FSC, DCLK,
DX, DR and MCLK.
FSC is a 8-kHz frame synchronization signal.
The DCLK is the clock signal to synchronize the data transfer on both data lines DX and
DR. The rising edge indicates the start of the bit while the falling edge is used to latch
the contents of the received data line DR. If the double clock rate is chosen (twice of the
transmission rate) the first rising edge indicates the start of a bit while the second falling
edge is used to latch the content of the data line.
The data rate of the interface can vary from 64 kbit/s to 4.096 Mbit/s. A frame may
consist of up to 64 time-slots of 8 bits each. The last 6 bits of TSCR (Time Slot
Configuration Register) indicate the selected time-slot from 0 to 63. If a 16-bit mode
(linear mode) is chosen, the lowest data rate is 128 kbit/s and the time-slot must be set
to an even number.
The pin AD/MCLK is a system clock synchronized with FSC (necessary to synchronize
internal PLL).
Figure 33 shows the timing of a serial data interface (256 kbit/s with single clock rate).
Figure 33
Serial Data Interface Timing
Semiconductor Group
71
Functional Description
2.4
Test Functions
The ARCOFI provides several test and diagnostic functions which can be grouped as
follows:
● All programmable configuration registers and coefficient RAM-locations are readable
● Digital loop via PCM-register (DLP)
● Digital loop via signal processor (DLS)
● Digital loop via noise shaper (DLN)
● Analog loop via analog front end (ALF)
● Analog loop via converter (ALC)
● Analog loop via noise shaper (ALN)
● Analog loop via Z-sidetone (ALZ); sidetone gain stage GZ must be enabled
(PFCR.GZ = 1) and sidetone gain must be programmed with 0 dB; depending on the
VDM-bit setting (DFICR) an addition to the incoming voice signal is possible
● Analog loop via digital interface (ALI).
Semiconductor Group
72
Operational Description
3
Operational Description
3.1
Reset
After a RESET (internal power-on reset, hardware reset at pin RS or software reset via
XOP_E) the pins SA to SD are programmed as inputs. All other output pins are in
high-impedance state (HOP/HON, LSP/LSN, VREF, PZ1, PZ2, DU/DX).
Note: After a Reset (only TE and Non-TE mode) the coefficient RAM-locations have
defined reset values.
The defined reset values of the ARCOFI-registers are listed below:
Register
Value after
RESET [hex]
Meaning
CMDR
GCR
BF
00
DFICR
F0
PFCR
00
TGCR
00
TGSR
ATCR
00
00
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
ARCR
00
TFCR
00
–
–
–
–
–
–
Semiconductor Group
No operation (NOP)
Speakerphone disabled (incl. AGCX and AGCR)
Disable voice transmit
IOM-2 channel 0 selected (IOM-2 TE-mode)
Power-down mode
IOM-2 two chip mode (IOM-2 TE-mode)
A-Law
SA to SD programmed as inputs
PCM-mode; receive voice blocked
Programmable digital gain disabled
Programmable sidetone gain disabled
Correction filters disabled
50-Hz receive HP active
50-Hz transmit HP active
Tone generator inactive
Control generator inactive
No tone generator connection to any signal path
Microphone amplifier is in power-down mode
Reference voltage buffer is in power-down mode
Pins MIP1/MIN1 are directed to the microphone
amplifier AMI
Earpiece amplifier AHO is in power-down mode
Loudspeaker amplifier ALS is in power-down
mode
IOM-2 handshake procedure enabled
No internal speakerphone status signals are
directed to PZ1/PZ2
Analog test mode disabled
Digital test mode disabled
73
Operational Description
Defined reset values of the ARCOFI-registers (cont’d)
Register
Value after
RESET [hex]
Meaning
SDICR
00
TSCR
XCR
CRAM
00
00
00
–
–
–
–
–
–
3.2
Single clock rate (DCLK) is enabled
DX and SDX are configured as open drain outputs
Master clock rate is 512 kHz
Time-slot 0 (SDI) is selected
AHO and ALS are in the differential mode
All locations (TE and Non-TE)
Initialization
During initialization a subset of configuration registers and coefficient RAM-locations has
to be programmed to set the configuration parameters according to the application and
desired features.
Configuration Registers:
Register
Bit
Effect
GCR
SP
AGCX
AGCR
EVX
SLOT
PU
CAM
LAW
SA-SD
VDM
GX
GR
GZ
FX
FR
DHPR
DHPX
TG
DT
ETF
CG
BT
BM
SM
Speakerphone ON/OFF
TX-automatic gain control (only if GCR.SP = 1)
RX-automatic gain control (only if GCR.SP = 1)
Enable voice transmit
IOM-2 slot select
Power-up/down mode
IOM-2 address mode
A-Law/µ-Law
PCI-port configuration
Voice data manipulation
TX digital gain
RX digital gain
Sidetone gain
TX-frequency correction filter
RX-frequency correction filter
Disable high-pass (50 Hz) receive
Disable high-pass (50 Hz) transmit
Tone generator
Dual tone mode
Enable tone filter
Control generator
Beat tone generator
Beat mode
Stop mode
DFICR
PFCR
TGCR
Semiconductor Group
Restricted to
74
IOM-2 TE
IOM-2
IOM-2 TE
Operational Description
Configuration Registers (cont’d)
Register
TGSR
ATCR
ARCR
TFCR
SDICR
TSCR
XCR
Bit
Effect
Restricted to
SQTR
PM
TRL
TRR
DTMF
TRX
MIC
EVREF
AIMX
HOC
CME
LSC
DHS
EPZST
ALTF
DLTF
EPP0
EPP1
DCE
MCLKR
TS
DHOP
DHON
DLSP
DLSN
Square/trapezoid shaped signal
Piezo mode
Tone ringing via loudspeaker
Tone ringing in receive direction
DTMF-signal in transmit direction
Tone ringing in transmit direction
Microphone amplifier control
Enable 2.4 V reference voltage at pin VREF
Analog input multiplexer
Handset amplifier control
Controlled monitoring
Loudspeaker amplifier control
Disable IOM-2 handshake procedure
PZ1/PZ2 as speakerphone status output
Analog Loops and test functions
Digital Loops and test functions
Enable push/pull (DX)
Enable push/pull (SDX)
Double clock enable
Master clock rate
Time slot select
Disable HOP (tristate)
Disable HON (tristate)
Disable LSP (tristate)
Disable LSN (tristate)
IOM-2
SDI
SDI
SDI
SDI
SDI
Note: Before accessing the ARCOFI PCI (IOM-2 TE-mode) interface, a GCR-write
command (SOP_0 or SOP_F) has to be sent.
Coefficient RAM-locations:
Mnemonic
# of Bytes
Effect
COP_0: Tone generator parameter set 1
F1
G1
GD1
T1
2
1
1
2
2
Semiconductor Group
Tone generator frequency
Tone generator amplitude
Trapezoid generator amplitude
Beat tone time
not used
75
Operational Description
Coefficient RAM-locations (cont’d)
Mnemonic
# of Bytes
Effect
COP_1: Tone generator parameter set 2; tone generator level adjustment
F2
G2
GD2
T2
GTR
GTX
2
1
1
2
1
1
Tone generator frequency
Tone generator amplitude
Trapezoid generator amplitude
Beat tone time span
Level adjustment for receive path
Level adjustment for transmit path
COP_2: Tone generator parameter set 3;
Parameter set for the DTMF-generator (TGSR.DTMF = 1)
F3
G3
GD3
T3
FD
2
1
1
2
2
Tone generator frequency
Tone generator amplitude
Trapezoid generator amplitude
Beat tone time span
Dual tone frequency
COP_3: Tone filter
K
A1
A2
GE
1
1
1
1
Attenuation of the stop-band
Center frequency
Bandwidth
Saturation amplification
COP_4: Control generator
TON
TOFF
2
2
Turn-on period of the tone generator
Turn-off period of the tone generator
COP_5: Receive and transmit gain
GX
GR
2
2
4
Transmit gain
Receive gain
Not used
COP_6: Sidetone gain
GZ
2
2
Sidetone gain
not used
COP_7/COP_8: Transmit correction filter
FX
12
Semiconductor Group
Transmit correction filter coefficients
76
Operational Description
Coefficient RAM-locations (cont’d)
Mnemonic
# of Bytes
Effect
COP_8/COP_9: Receive correction filter
FR
12
Receive correction filter coefficients
COP_A: Parameter set for transmit and receive speech comparator
Parameter set for speakerphone control unit
GAE
GLE
ATT
ETAE
ETLE
TW
DS
SW
1
1
1
1
1
1
1
1
Gain of acoustic echo
Gain of line echo
Attenuation programmed in GHR or GHX
Echo time (acoustic side)
Echo time (line side)
Wait time
Decay speed
Switching time
COP_B: Parameter set for transmit and receive speech comparator
GDSAE
PDSAE
GDNAE
PDNAE
GDSLE
PDSLE
GDNLE
PDNLE
1
1
1
1
1
1
1
1
Reserve when speech is detected (acoustic side)
Peak decrement when speech is detected (acoustic side)
Reserve when noise is detected (acoustic side)
Peak decrement when noise is detected (acoustic side)
Reserve when speech is detected (line side)
Peak decrement when speech is detected (line side)
Reserve when noise is detected (line side)
Peak decrement when noise is detected (line side)
COP_C: Parameter set for transmit and receive speech detector
LIM
OFFX
OFFR
LP2LX
LP2LR
LP1X
LP1R
1
1
1
1
1
1
1
1
Semiconductor Group
Starting level of the logarithmic amplifiers
Level offset up to detected noise (transmit)
Level offset up to detected noise (receive)
Limitation for LP2 (transmit)
Limitation for LP2 (receive)
Time constant LP1 (transmit)
Time constant LP1 (receive)
not used
77
Operational Description
Coefficient RAM-locations (cont’d)
Mnemonic
# of Bytes
Effect
COP_D: Parameter set for receive and transmit speech detector
PDSX
PDNX
LP2SX
LP2NX
PDSR
PDNR
LP2SR
LP2NR
1
1
1
1
1
1
1
1
Time constant PD for signal (transmit)
Time constant PD for noise (transmit)
Time constant LP2 for signal (transmit)
Time constant LP2 for noise (transmit)
Time constant PD for signal (receive)
Time constant PD for noise (receive)
Time constant LP2 for signal (receive)
Time constant LP2 for noise (receive)
COP_E: Parameter set for transmit AGC
LGAX
COMX
AGX
TMHX
TMLX
NOISX
1
1
1
1
1
1
1
1
Loudness gain adjustment
Compare level rel. to max. PCM-value
Gain range of automatic control
Settling time constant for lower levels
Settling time constant for higher levels
Threshold for AGC-reduction by background noise
not used
not used
COP_F: Parameter set for receive AGC
LGAR
COMR
AAR
AGR
TIMHR
TIMLR
NOISR
1
1
1
1
1
1
1
1
Semiconductor Group
Loudness gain adjustment
Compare level rel. to max. PCM-value
Attenuation range of automatic control
Gain range of automatic control
Settling time constant for lower levels
Settling time constant for higher levels
Threshold for AGC-reduction by background noise
not used
78
Operational Description
3.3
ARCOFI® Operating Modes
The most currently used ARCOFI-operating modes are documented in the following
table. The 12 ARCOFI-configuration registers have enough build-in flexibility to
accommodate an extensive set of user calling procedures.
The following operating mode description table is not exhaustive but should be used as
an example of possible functions performed by the ARCOFI.
State
Description
POR
Power-on reset: when power is supplied to the ARCOFI an
internal power-on reset is generated. In addition a hardware reset
via an RC-network connected to input pin RS will force all
ARCOFI internal registers to default values. The
ARCOFI-registers reset state is described in section 3.1 and 4.
STAND BY
The system microprocessor can initialize the ARCOFI via the
IOM-2 or the SCI-bus with a different set of filter and
configuration values. Whilst remaining in power-down
(GCR.PU = 0) a new set of filter coefficients and configuration
bits can be loaded in the ARCOFI.
HANDSET
The system MPU detects activity from the hookswitch or from the
keyboard. The ARCOFI can be placed in HANDSET state where
all handset l/O are enabled (AMI & AHO activated).
RINGING
The system MPU detects an incoming call, the ARCOFI can be
placed in a RINGING state by activating the tone ringer via
TGCR/TGSR and configuring the ARCOFI such that either the
LSP/LSN-output or the piezo output (pins PZ1/PZ2) are enabled.
An emergency ringing is also implemented. In this mode, only the
tone ringer and the loudspeaker amplifier are active (AMI- and
AHO-amplifier are disabled by the user). The tone ringer signal is
directly switched to the loudspeaker amplifier ALS.
DTMF
All audio inputs can be disabled by forcing the AMI-amplifier
(ATCR) to power-down. DTMF tones are generated with the tone
generator and are output to the transmit path.
PULSE DIAL
Handset audio path can be enabled by forcing a HANDSET
mode. A single tone can be superimposed into the audio receive
path so as to provide audible feedback when dialling.
LOUD HEARING
(MONITORING)
The handset l/O and the loudspeaker outputs LSP/LSN are
active (ATCR & ARCR).
Semiconductor Group
79
Operational Description
Operating mode description table (cont’d)
State
Description
SPEAKERPHONE
The handset audio I/O's are disabled. The hands-free
microphone input and loudspeaker outputs LSP/LSN are
activated by configuring ATCR & ARCR. The ARCOFI must be
set to the speakerphone mode (GCR.SP = 1).
MUTE
The ARCOFI can be placed in a MUTE state by powering down
the AMI. In handset mode the outputs HOP/HON remain enabled
while in speakerphone mode the outputs LSP/LSN are enabled.
All other analog l/O's being disabled.
FEATURE TONE
A single tone can be superimposed to the incoming PCM-voice
signal. Applications requiring system function audible feedback
are therefore made possible.
Semiconductor Group
80
Operational Description
3.4
IOM®-2 Interface Protocol
The following description of the IOM-2 interface comprises all ARCOFI relevant functions
in the terminal and non-terminal mode (see IOM-2 interface specification for general
information).
Note: Channels IC1 & IC2 are only available in the IOM-2 TE-mode. MON-channel
means MON1-channel in the IOM-2 TE-mode.
3.4.1
B- and IC-Channels
The ARCOFI can receive and transmit voice data in the IOM-2 B1- & B2-channels as
well as in the IC1- & IC2-intercommunication channels located in IOM-2 channels 0 and
1 respectively. The voice/data channel allocation is programmable via the
ARCOFI-channel select bit SLOT in the GCR-register. B1 or B2 or respectively IC1 or
IC2 can be programmed by use of the RCM-bit in the CMDR-register.
The IC1- and IC2-intercommunication channels can be used in the terminal for local data
communication (e.g. answering machine). This makes post-processing of voice/data
information possible (e.g. data encryption).
3.4.2
Monitor Channel
All programming data required by the ARCOFI including coefficients are transmitted
exclusively in the MON-time-slot of the IOM-2 channel. The MON-channel allows a point
to multi-point access where the layer-2 component acts as the master to program
devices like the ARCOFI. Each programmable device is accessed by sending a specific
address byte at the start of each SOP- or COP-command stream. Before executing a
command, the programmable device compares the received address byte with its own
address. The latter consists of 8 bits whose 4th MSB-bit must correspond to the AD-wire
(AD/MCLK pin) strapped IOM-2 address.
3.4.2.1 MON-Channel Data Structure
The data to control and program the ARCOFI are transferred in the MON1-channel via
the IOM-2 interface by a procedure utilizing read/write registers in the ARCOFI.
The messages transmitted in the monitor channel may have different kinds of data
structures. Therefore, the first byte of the message is used to indicate the data structure
(first four bits).
Semiconductor Group
81
Operational Description
Identification Command
In order to be able to identify unambiguously different devices by software, the following
identification command is used:
DD 1st byte value
1
0
1
X
0
0
0
0
DD 2nd byte value
0
0
0
0
0
0
0
0
The ARCOFI responds to this DD-identification sequence by sending a DU identification
sequence:
DU 1st byte value
1
0
DU 2nd byte value
1
0
1
X
0
0
0
0
DESIGN
X:
logical 0 (active low):
logical 1 (passive high):
AD = 0 (A-chip)
AD = 1 (B-chip)
DESIGN:
six bit code, specific for each device in order to identify differences in
operation
e.g.
000000
000010
000100
ARCOFI
ARCOFI-SP
ARCOFI-SP
PSB 2160
PSB 2165
PSB 2163
This identification sequence is usually done once, when the terminal is connected for the
first time.This function is used so that the software can distinguish between different
possible hardware configurations. However this sequence is not compulsory.
Programming Sequence
An ARCOFI-programming sequence is characterized by a “1” being sent in the
LSB-nibble of the first incoming identification code.
DD 1st byte value
DD 2nd byte value
1
0
1
X
0
0
0
1
COP_X, SOP_X, XOP_X
All programmed configurations and coefficients can be read back when issuing an
appropriate CMDR read (CMDR.R/W = 1). The ARCOFI responds by sending an IOM-2
specific address byte identifying the chip followed by the requested data.
Semiconductor Group
82
Operational Description
3.4.2.2 MON-Transfer Protocol
The transfer of a stream of commands in the MON-channel is regulated by a handshake
protocol mechanism implemented by two bits MX and MR in the fourth slot of the IOM-2
channel. The procedure is as follows (figure 34):
Figure 34
Monitor Channel Handshake Procedure
Semiconductor Group
83
Operational Description
Monitor transfer protocol rules:
● A pair of MX and MR in the inactive state for two or
●
●
●
●
●
●
●
●
more consecutive frames
indicates an idle state or an end of transmission (EOM).
A command stream initiated by a transmitter in the MON-slot is accompanied by an
activated downstream MX-bit.
The receiver acknowledges a received byte by toggling the upstream MR-bit from
inactive to active in the subsequent IOM-2 frame for at least one frame.
The transmitter indicates a new byte in the MON-slot by the transition of the MX-bit
from the active to the inactive state. The MX-bit returns to the active state after one
frame. Two frames with the MX-bit in the inactive state indicate the end of
transmission.
The receiver acknowledges each new byte by a similar one frame transition of the
MR-bit to the inactive state. Two frames with the MR-bit set to inactive indicate a
receiver request for abort.
The transmitter can delay a transmission sequence by sending the same byte
continuously. In that case the MX-bit remains active in the IOM-2 frame following the
first byte occurrence.
Delaying a transmission sequence is only possible while the receiver MR-bit and the
transmitter MX-bit are active.
Since the receiver is able to receive the MON-slot data at least twice (in two
consecutive frames), the receiver waits for the reception of two successive identical
bytes.
To control this handshake procedure a collision detection mechanism is implemented
in the transmitter. This is done by making a collision check per bit on the transmitted
MON-data.
Semiconductor Group
84
Operational Description
3.4.2.3 Implementation of the MON-Channel Protocol
The MON-receiver has the following features:
● Transparent interface between IOM-2 interface and any device internal block (sink)
with respect to handshake procedure, i.e. any acknowledge, EOM, abort or request
for abort is conveyed transparently through the receiver.
Figure 35 shows the state diagram of the MON-receiver. The following signals are used:
MR:
MX:
LL:
MR-bit sent by the receiver
MX-bit received
Last two bytes were identical
Figure 35
State Diagram of the Monitor Receiver
Semiconductor Group
85
Operational Description
The MON-transmitter has the following features:
● Transparent interface between IOM-2 interface and any device internal block (source)
with respect to handshake procedure, i.e. any acknowledge, abort, request for abort
is conveyed transparently through the transmitter.
Figure 36 shows the state diagram of the MON-transmitter. The following signals are
used:
MR:
MX:
LL:
RQT:
EOM:
MR-bit received
MX-bit transmitted
Last two bytes were identical
Request transmission
End of transmission
Figure 36
State Diagram of the Monitor Transmitter
Semiconductor Group
86
Operational Description
3.4.3
Command/Indication Channel 1 (TE-mode)
The C/l-channel bits are represented so that the first bit transmitted/received appears on
the left. The data presented to the four peripheral control interface (PCI) pins SA to SD
are transparently routed to the C/l IOM-2 channel 1. Pins SA to SD can be configured
individually as input or output and the information sent to the pins SA to SD or coming
from them will appear respectively in the DD or DU C/l IOM-2 channel 1.
In case a reset has been asserted, the SA- to SD-pins are programmed as input,
however the SA- to SD-values are not switched to the DU C/I1-channel unless a write
command (except NOP) is issued.
The mapping of the peripheral control interface (PCI) pins SA to SD into the six
C/I1-channel bits depends on the hardwired AD-address (see section 3.7) as follows.
AD = 1
7
DD and DU
–
2
–
SB
1
0
SA
SD
SC
MR MX
–
–
–
MR MX
SA
–
–
MR MX
AD = 0 (GCR.CAM = 0; two chip mode)
DD and DU
SD
SC
–
AD = 0 (GCR.CAM = 1; one chip mode)
DD and DU
Semiconductor Group
SD
SC
SB
87
Operational Description
C/I1-Channel (Signaling) Bit Allocation Table:
CAM
X
AD
DD-C/I1
X
7 6 5 4 3 2 7 6 5 4 3 2
X X X X X X H H H H H H after reset
0
0
0
0
0
0
X
S
D
S
D
0
0
X
DU-C/I1
S S
X X X X X D C H H H H PCI-pins as inputs
S
C X X X X H H H H H H PCI-pins as outputs
S
X X X X X H C H H H H PCI-pin SC as input
PCI-pin SD as output
S
S
C X X X X D H H H H H PCI-pin SD as input
PCI-pin SC as output
X
1
X
1
X
1
X X X X X
S S S
X X B A D
S
X X X A X
X
1
S
S
X X B X D
1
0
1
0
1
0
X
S
D
S
D
0
S
X C X
1
X:
H:
X X
S S
C B
S
X B
S S S S
X H H B A D C PCI-pins as inputs
S
C H H H H H H PCI-pins as outputs
S
S
S
C H H B H D H SB and SD as inputs
SA and SC as outputs
S
S
X H H H A H C SA and SC as inputs
SB and SD as outputs
S S S S
X X X D C B A H H PCI-pins as inputs
S
A X X H H H H H H PCI-pins as outputs
S
S
X X X H C H A H H SA and SC as inputs
SB and SD as outputs
S
S
S
A X X D H B H H H SB and SD as inputs
SA and SC as outputs
don’t care
passive high
Semiconductor Group
PCI-Configuration
88
Operational Description
3.5
ARCOFI® Voice/Data Manipulation (VDM)
The ARCOFI offers several possibilities of voice/data manipulation for special
applications.
According to the manipulation mode chosen, the byte B1 or B2 (or IC1 or IC2 in the
IOM-2 TE-mode) can be output via the handset channel and/or the loudspeaker channel.
The following tables gives an overview of the different voice/data manipulation modes.
● PCM-mode or normal mode (DFICR.VDM = 000X):
DFICR.VDM
AD Pin
(IOM-2)
1)
CMDR.RCM
(IOM-2)
GCR.SLOT
(IOM-2)
Receive
Channel
Transmit
Channel
0000
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
–
–
–
–
–
–
–
–
B1
IC1
B2
IC2
B2
IC2
B1
IC1
0001
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
B1
IC1
B2
IC2
B2
IC2
B1
IC1
B1
IC1
B2
IC2
B2
IC2
B1
IC1
● Linear mode (DFICR.VDM = 010X):
This mode exists only in the SDI mode (in the programmed and the following channel)
or in the IOM-2 one chip mode (GCR.CAM = 1). The two voice/data channels B1 and B2
(or IC1 and IC2 in the IOM-2 TE-mode) are connected to one 16-bit linear channel (2s
complement).
DFICR.VDM
AD Pin
(IOM-2)
CMDR.RCM
(IOM-2)
GCR.SLOT
(IOM-2)
Receive
Channel
Transmit
Channel
0100
–
–
–
–
0
1
–
–
B1 & B2
IC1 & IC2
0101
–
–
–
–
0
1
B1 & B2
IC1 & IC2
B1 & B2
IC1 & IC2
1)
This table is given for the IOM-2 two chip mode (GCR.CAM = 0).
Semiconductor Group
89
Operational Description
B1&B2 (IC1&IC2) means B1 (IC1) byte followed by B2 (IC2) byte (totally 16 bits).
● Three party conferencing (DFICR.VDM = 100X):
This mode is available only in the SDI-mode (in the programmed and the following
channel) or in the IOM-2 one chip mode (GCR.CAM = 1).
DFICR.VDM
AD Pin
(IOM-2)
CMDR.RCM
(IOM-2)
GCR.SLOT
(IOM-2)
Receive
Channel
Transmit
Channel
1000
–
–
–
–
0
1
B1 + B2
IC1 + IC2
B1, B2
IC1, IC2
B1 + B2 (IC1 + IC2) means the B1 (IC1) and the B2 (IC2) byte are added together (on
8 bits).
B1, B2 (IC1, IC2) means B1 (IC1) and B2 (IC2) byte have the same information.
● Voice monitoring mode (DFICR.VDM = 1100):
This mode is available only in the SDI-mode or in the IOM-2 one chip mode
(GCR.CAM = 1). The monitoring chip and the transmission chip must be strapped to a
different hardware address (IOM-2: AD/MCLK pin).
The active DU-voice channel of the monitoring chip must be set in the Hi Z-mode
(GCR.EVX = 0).
The PCI-port of both chips must be set in the compatible configurations to avoid collision
problems in the DU-C/I1 channel.
DFICR.VDM
AD Pin
(IOM-2)
CMDR.RCM
(IOM-2)
GCR.SLOT
(IOM-2)
Receive
Channel
Transmit
Channel
1100
1/0
1/0
1/0
1/0
0
0
1
1
0
1
0
1
B1D + B1U
IC1D + IC1U
B2D + B2U
IC2D + IC2U
B1
IC1
B2
IC2
Explanations:
–
B1/B2
IC1/IC2
B1D/B2D
B1U/B2U
IC1D/IC2D
IC1U/IC2U
Semiconductor Group
no signal
voice channels
IOM-2 intercommunication channels
IOM-2 voice channels (downstream)
IOM-2 voice channels (upstream)
IOM-2 intercommunication channels (downstream)
IOM-2 intercommunication channels (upstream)
90
Operational Description
Figure 37
Configuration of the IOM®-2 TE-Monitoring Mode
Semiconductor Group
91
Register Description
4
Detailed Register Description
The following section describes the various ARCOFI-registers and coefficient
RAM-locations accessible from the terminal equipment microcontroller via the IOM-2
bus or via the serial controller interface (SCI).
A summary of the 12 registers located in the ADI-block is presented below followed by
a detailed description of the register content.
Command Register (CMDR)
7
CMDR
0
R/W
RCM
CMD5
CMD4
CMD3
CMD2
CMD1
CMD0
General Configuration Register (GCR)
7
GCR
SP
0
AGCX
AGCR
EVX
SLOT
PU
CAM
LAW
Data Format and Interface Configuration Register (DFICR)
7
DFICR
SD
0
SC
SB
SA
VDM
Programmable Filter Configuration Register (PFCR)
7
PFCR
GX
0
GR
GZ
FX
0
FR
DHPR
DHPX
Tone Generator Configuration Register (TGCR)
7
TGCR
TG
0
DT
ETF
CG
BT
BM
SM
SQTR
Tone Generator Switch Register (TGSR)
7
TGSR
PM
0
TRL
Semiconductor Group
0
TRR
92
DTMF
TRX
0
0
Register Description
AFE Transmit Configuration Register (ATCR)
7
0
ATCR
MIC
EVREF 0
AIMX
AFE Receive Configuration Register (ARCR)
7
0
ARCR
HOC
CME
LSC
Test Function Configuration Register (TFCR)
7
TFCR
0
0
EPZST
ALTF
DLTF
SDI-Configuration Register (SDICR); only available in SDI-mode
7
SDICR
0
0
0
EPP1
EPP0
DCE
MCLKR
Time-Slot Configuration Register (TSCR); only available in SDI-mode
7
TSCR
0
0
0
TS
Extended Configuration Register (XCR)
7
XCR
PGCR
0
PGCX
RAAR
OBS
DHOP
DHON
DLSP
DLSN
Test Mode Register (TMR)
7
TMR
Semiconductor Group
0
TM
0
0
93
0
0
0
Register Description
4.1
Command Register (CMDR)
Value after reset: BFH
7
CMDR
0
R/W
RCM
CMD5
CMD4
CMD3
CMD2
CMD1
CMD0
R/W
0: writing to configuration registers or to coefficient RAM
1: reading from configuration registers or from coefficient RAM
RCM
Reverse Channel Mode (if R/W = 0)
0: receive and transmit in B1 (or IC1 in TE-mode)
1: receive and transmit in B2 (or IC2 in TE-mode)
For the IOM-2 two chip mode (GCR.CAM = 0), when pin AD is strapped
to VSS the above applies. When pin AD is strapped to VDD,
RCM operates in the reverse order.
CMDx
Address to internal programmable locations
CMD 5 4 3 2 1 0
0 0 X X X X
code reserved
0 1 X X X X
status operation (SOP)
1 0 X X X X
coefficient operation (COP)
1 1 X X X X
extended operation (XOP)
Coding of status operations (SOP):
Bit 3
2
1
0
CMD
Name
Status
CMD
Seq. Len.
CMD Sequence
Description
0
0
0
0
0
0
0
0
1
1
1
1
1
1
0
0
0
0
1
1
1
1
0
0
0
1
1
1
0
0
1
1
0
0
1
1
0
0
1
0
1
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
SOP_0
SOP_1
SOP_2
SOP_3
SOP_4
SOP_5
SOP_6
SOP_7
SOP_8
SOP_9
SOP_A
SOP_D
SOP_E
SOP_F
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R
R/W
R/W
2
2
2
2
2
2
2
2
2
2
2
2
2
9
<GCR>
<DFICR
<PFCR>
<TGCR>
<TGSR>
<ATCR>
<ARCR>
<TFCR>
<SDICR>
<TSCR>
<XCR>
<IDENT>1)
<TMR>
<TFCR>..<GCR>
1) See also 4.12.
Semiconductor Group
94
Register Description
Coding of coefficient operations (COP):
Bit 3 2
1
0
CMD
Name
Status
CMD CMD
Seq. Sequence
Len. Description
Comments
0
0
0
0
COP_0
R/W
9
Tone generator 1
0
0
0
1
COP_1
R/W
9
0
0
1
0
COP_2
9
R/W
0
0
0
1
1
0
1
0
COP_3
COP_4
R/W
R/W
5
5
0
1
0
1
COP_5
R/W
9
0
1
1
0
COP_6
R/W
5
0
1
1
0
1
0
1
0
COP_7
COP_8
R/W
R/W
9
9
1
1
1
1
1
1
1
1
0
1
0
1
0
1
0
1
1
0
0
1
1
0
0
0
1
1
1
1
COP_9
COP_A
COP_B
COP_C
COP_D
COP_E
COP_F
<F1> <F1> <G1> <GD1>
<T1> <T1> <..> <..>
<F2> <F2> <G2> <GD2>
<T2> <T2>
<GTR> <GTX>
<F3> <F3> <G3> <GD3>
<T3> <T3>
<FD> <FD>
<K> <A1> <A2> <GE>
<TON> <TON>
<TOFF> <TOFF>
<GX> <GX>
<GR> <GR>
<..> <..> <..> <..>
<GZ> <GZ>
<..> <..>
<FX1>..<FX8>
<FX9>..<FX12>
<FR1>..<FR4>
<FR5>..<FR12>
<SP1>..<SP8>
<SP9>..<SP16>
<SP17>..<SP24>
<SP25>..<SP32>
<AGCX1>..<AGCX8>
<AGCR1>..<AGCR8>
9
9
9
9
9
9
9
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Tone generator 2
Additional TG-gain
Tone generator 3
Dual tone frequency
Tone filter
Control generator
Transmit gain
Receive gain
Sidetone gain
Correction filter FX
Correction filter FR
Coefficients for
Speakerphone
AGC-transmit
AGC-receive
Coding of extended operations (XOP):
Bit 3
2
1
0
CMD
Name
Status
CMD
Seq. Len.
Comments
0
0
1
0
0
1
0
0
0
0
1
1
XOP_0
XOP_1
XOP_D
W
W
W
1
1
1
1
1
1
1
1
1
0
1
XOP_E
XOP_F
W
R/W
1
1
Power-down mode
Power-up mode
DD/DU-voice channel
swap (toggle function)
Software reset
Normal operation (NOP)
Semiconductor Group
95
Register Description
4.2
General Configuration Register (GCR)
Value after reset: 00H
7
GCR
SP
0
AGCX
AGCR
EVX
SLOT
PU
CAM
LAW
SP
Speakerphone
0: speakerphone support disabled
1: speakerphone support enabled
AGCX
Automatic Gain Control Transmit
0: automatic gain control disabled
1: automatic gain control enabled; only if speakerphone support
is enabled (SP = 1)
AGCR
Automatic Gain Control Receive
0: automatic gain control disabled
1: automatic gain control enabled; only if speakerphone support
is enabled (SP = 1)
EVX
Enable Voice Transmit
0: disable transmit voice data
1: enable transmit voice data (if GCR.PU = 0, idle code is transmitted)
SLOT
IOM-2 Slot Select (IOM-2 TE-mode only)
0: bearer channels in IOM-channel 0
1: bearer channels in IOM-channel 1
PU
Power-Up
0: the ARCOFI is placed in standby mode (power-down); all registers
and coefficient RAM contents are saved and all interface functions
are available
1: the ARCOFI is in a normal operating mode (power-up)
This bit can be directly accessed by the XOP_0/XOP_1 operations.
CAM
Chip Address Mode (IOM-2 mode only)
0: two ARCOFIs are connected to the IOM-2 bus
1: only one ARCOFI is connected to the IOM-2 bus
LAW
Coding Law
0: A-Law enabled
1: µ-Law enabled
Semiconductor Group
96
Register Description
4.3
Data Format and Interface Configuration Register (DFICR)
Value after reset: F0H
7
DFICR
0
SD
SC
SB
SA
VDM
SD-SA
Signaling I/O (PCI-interface; only available in IOM-2 TE-mode)
0: Sx pin programmed as output (x: D, C, B or A)
1: Sx pin programmed as input (x: D, C, B or A)
VDM
Voice Data Manipulation
Bit 3 2
1
0
Receive
Transmit
Description
Voice Channel Voice Channel
0
0
0
0
–
B1
Transmit only
0
0
0
1
B1
B1
Transfer mode
0
1
0
0
–
B1 & B2
16-bit transmit only
0
1
0
1
B1 & B2
B1 & B2
16-bit transfer mode
1
0
0
0
B1 + B2
B1, B2
Conferencing mode
1
1
0
0
B1D + B1U
B1
Monitoring mode
– no signal
Note: In this table above the voice channels indicated are only examples.
Other combinations of B1 and B2 (IC1 and IC2 in IOM-2 TE-mode)
are possible and a complete description is given in section 3.5.
Semiconductor Group
97
Register Description
4.4
Programmable Filter Configuration Register (PFCR)
Value after reset: 00H
7
PFCR
GX
0
GR
GZ
FX
0
FR
DHPR
GX
Transmit Gain
0: gain set to 0 dB
1: gain coefficients loaded from coefficient RAM (CRAM)
GR
Receive Gain
0: gain set to 0 dB
1: gain coefficients loaded from CRAM
GZ
Sidetone Gain
0: gain set to – ∞ dB
1: gain coefficients loaded from CRAM
FX
Transmit Frequency Correction Filter
0: filter is by-passed
1: filter coefficients loaded from CRAM
FR
Receive Frequency Correction Filter
0: filter is by-passed
1: filter coefficients loaded from CRAM
DHPR
Disable High-Pass Receive (50/60 Hz filter)
0: filter enabled
1: filter disabled
DHPX
Disable High-Pass Transmit (50/60 Hz filter)
0: filter enabled
1: filter disabled
Semiconductor Group
98
DHPX
Register Description
4.5
Tone Generator Configuration Register (TGCR)
Value after reset: 00H
7
TGCR
TG
0
DT
ETF
CG
BT
BM
SM
SQTR
TG
Tone Generator
0: tone generator is disabled (if CG = 0)
1: tone generator is enabled; frequency and gain coefficients loaded
from CRAM; CG has priority over TG
DT
Dual Tone Mode (DTMF)
0: second trapezoid generator is disabled
1: second trapezoid generator is enabled; the output signal is added
to the S/T signal generator (only if TGSR.DTMF = 0)
ETF
Enable Tone Filter
0: tone filter is by-passed
1: tone filter is enabled; filter coefficients loaded from CRAM
CG
Control Generator
0: control generator is disabled
1: control generator is enabled; time coefficients loaded from CRAM
(tone generator is activated independently of TG-bit setting)
BT
Beat Tone Generator
0: beat tone generator is disabled
1: beat tone generator is enabled; time coefficients loaded from CRAM
BM
Beat Mode
0: beat mode is disabled; two tone ring activated when BT-generator
is enabled
1: beat mode is enabled; three tone ring activated when BT-generator
is enabled (only if TGSR.DTMF = 0)
SM
Stop Mode
0: automatic stop mode is disabled
1: automatic stop mode is enabled; two and three tone ring gets
turned off after the sequence is completed
SQTR
Square/Trapezoid Waveform
0: trapezoid shaped signal is enabled (
only if tone ringing via loudspeaker and piezo mode is disabled:
TGSR.TRL = 0 & TGSR.PM = 0)
1: square-wave signal is enabled
Semiconductor Group
99
Register Description
4.6
Tone Generator Switch Register (TGSR)
Value after reset: 00H
7
TGSR
PM
0
TRL
0
TRR
DTMF
TRX
0
0
PM
Piezo Mode
0: ringing signal is not output to the piezo ring pins
1: ringing signal (square) is output to the piezo ring pins PZ1/PZ2
TRL
Tone Ringing via Loudspeaker
0: ringing signal is not output directly to the loudspeaker pins
1: ringing signal (square) is output directly to the loudspeaker pins
LSP/LSN
TRR
Tone Ringing Receive
0: tone generator for receive direction is disabled
1: tone generator for receive direction is enabled
DTMF
DTMF-Generator (transmit)
0: DTMF-generator for transmit direction is disabled
1: DTMF-generator for transmit direction is enabled
TRX
Tone Ringing Transmit
0: tone generator for transmit direction is disabled
1: tone generator for transmit direction is enabled
Semiconductor Group
100
Register Description
4.7
AFE-Transmit Configuration Register (ATCR)
Value after reset: 00H
7
0
ATCR
MIC
MIC
EVREF 0
AIMX
Microphone Control
Bit 7
6
5
4
Selected Mode
0
0
0
0
0
0
0
0
1
1
0
0
0
0
1
1
1
1
0
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
0
1
0
1
AMI is in power-down mode
0 dB amplification
6 dB amplification
12 dB amplification
18 dB amplification
24 dB amplification
30 dB amplification
36 dB amplification
42 dB amplification
AMI is in by-pass mode
EVREF
Enable VREF (2.4-V reference voltage)
0: VREF-buffer is enabled in function of bit GCR.PU (global power-up) and
ATCR/ARCR-programming
1: VREF-buffer and internal reference voltage generation are enabled
independently of the ARCOFI-configuration
AIMX
Analog Input Multiplexer
Bit 1
0
Selected Input
0
0
1
1
0
1
0
1
AMI is connected to the pins MIP1/MIN1 (differential input)
AMI is connected to the pins MIP2/MIN2 (differential input)
AMI is connected to the pin MI3 (single-ended input)
not used
Semiconductor Group
101
Register Description
4.8
AFE-Receive Configuration Register (ARCR)
Value after reset: 00H
7
0
ARCR
HOC
HOC
CME
LSC
Handset Output Control
Bit 7
6
5
Selected Mode
0
0
0
0
1
1
1
0
0
1
1
0
0
1
0
1
0
1
0
1
1
AHO is in power-down mode
2.5 dB amplification
– 3.5 dB amplification
– 9.5 dB amplification
– 15.5 dB amplification
– 21.5 dB amplification
AHO is in by-pass mode
CME
Control Monitoring Enable (GCR.SP = 1)
0: controlled monitoring disabled
1: controlled monitoring enabled (if transmit speech is detected,
ALS-programming is fixed to – 9.5 dB if LSC > – 9.5 dB)
LSC
Loudspeaker Output Control
Bit 3
2
1
0
Selected Mode
0
0
0
0
0
0
0
0
1
1
1
1
1
1
0
0
0
0
1
1
1
1
0
0
0
0
1
1
0
0
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
ALS is in power-down mode
11.5 dB amplification
8.5 dB amplification
5.5 dB amplification
2.5 dB amplification
– 0.5 dB amplification
– 3.5 dB amplification
– 6.5 dB amplification
– 9.5 dB amplification
– 12.5 dB amplification
– 15.5 dB amplification
– 18.5 dB amplification
– 21.5 dB amplification
ALS is in by-pass mode
Semiconductor Group
102
Register Description
4.9
Test Function Configuration Register (TFCR)
Value after Reset: 00H
7
TFCR
0
0
EPZST
ALTF
DLTF
EPZST
Enable PZ1/PZ2 (piezo pins) to output internal Status Information
0: PZ1/PZ2 are used as piezo port pins
1: PZ1/PZ2 are used to indicate internal speakerphone conditions
(PZ1: speakerphone idle condition; PZ2: speakerphone transmit
active)
Note: PM-bit must be set (TGSR.PM = 1)
ALTF
Analog Loop and Test Functions
DLTF
Bit 5
4
3
Test Function
0
0
0
0
1
0
0
1
1
0
0
1
0
1
0
NOT: No Test Mode
ALF: Analog Loop via Front End
ALC: Analog Loop via Converter
ALN: Analog Loop via Noise Shaper
ALI: Analog Loop via Interface
Digital Loop and Test Functions
Bit 2
1
0
Test Function
0
0
0
0
1
0
0
1
1
0
0
1
0
1
0
NOT: No Test Mode
IDR: Initialize DRAM
DLP: Digital Loop via PCM-Register
DLS: Digital Loop via Signal Processor
DLN: Digital Loop via Noise Shaper
Semiconductor Group
103
Register Description
4.10
SDI-Configuration Register (SDICR); SDI-mode only
Value after reset: 00H
7
SDICR
0
0
0
EPP1
EPP0
DCE
EPP0
Enable Push-Pull at pin DU/DX (SDI-mode only)
0: open drain enabled
1: push-pull enabled
EPP1
Enable Push-Pull at pin SDR/SDX (SDI-mode only)
0: open drain enabled
1: push-pull enabled
DCE
Double Clock Enable for DCLK (SDI-mode only)
0: single clock rate
1: double clock rate
MCLKR
Master Clock Rate (synchronized system clock)
Bit 2
1
0
MCLK-Clock Rate
0
0
0
512 kHz
0
0
1
1.536 MHz
0
1
0
2.048 MHz
0
1
1
4.096 MHz
1
X
X
16.384 MHz (test mode)
Semiconductor Group
104
MCLKR
Register Description
4.11
Time-Slot Configuration Register (TSCR); SDI-mode only
Value after Reset: 00H
7
TSCR
TS
0
0
0
TS
Time-Slot Selection
Bit 5
4
3
2
1
0
Time-Slot
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
Semiconductor Group
1
.
.
.
.
1
1
1
105
1
63
Register Description
4.12
Extended Configuration Register (XCR)
Value after reset: 00H
7
XCR
0
PGCR
PGCX
RAAR
OBS
DHOP
DHON
DLSP
DLSN
PGCR
Position of Gain Control Receive
0: in front of the speech detector
1: behind the speech detector
PGCX
Position of Gain Control Transmit
0: behind the speech detector
1: in front of the speech detector
RAAR
Read Automatic Attenuation Receive
0: disabled
1: enabled
OBS
Observation of internal modes (only for internal tests)
0: disabled
1: enabled
DHOP
Disable HOP-Amplifier
0: HOP-amplifier normal mode
1: Disable HOP-amplifier (power-down, output high impedance)
DHON
Disable HON-Amplifier
0: HON-amplifier normal mode
1: Disable HON-amplifier (power-down, output high impedance)
DLSP
Disable LSP-Amplifier
0: LSP-amplifier normal mode
1: Disable LSP-amplifier (power-down, output high impedance)
DLSN
Disable LSN-Amplifier
0: LSN-amplifier normal mode
1: Disable LSN-amplifier (power-down, output high impedance)
Note: XCR.RAAR = 1, SOP_D (read) monitors the magnitude of the gain-cell AGCR
instead of IDENT.
Semiconductor Group
106
Register Description
4.13
Test Mode Register (TMR)
Value after reset: 00H
7
0
TMR
TM
TM
0
0
Test Mode (only for internal tests)
000: normal mode
Semiconductor Group
107
0
0
0
Electrical Characteristics
5
Electrical Characteristics
Absolute Maximum Ratings
Parameter
Symbol
Ambient temperature under bias
TA1)
TSTG
VS
Vmax
Storage temperature
Voltage on any pin with respect to ground
Maximum voltage on any pin
1)
Limit Values
Unit
– 25 to 80
˚C
– 65 to125
˚C
– 0.3 to VDD + 0.3
V
7
V
Reduced performance
ESD-integrity (according MIL-Std 883D, method 3015.7): 1000 V
exception: The pins #14, #16, #17 and #18 are not protected against voltage stress
> 630 V (versus VSSx, x = A, D, P). The output performance prohibits the use of
adequate protective structures.
Note: Stresses above those listed here may cause permanent damage to the device.
Exposure to absolute maximum ratings conditions for extended periods may affect
device reliability.
DC-Characteristics
VDD/VDDP = 5 V ± 5 %; VSSD/VSSA/VSSP = 0 V; TA = 0 to 70 ˚C
Parameter
Input leakage current
H-input level
(except pins SCLK,
MCLK, DCLK)
Symbol
IIL
VIH1
Limit Values
Unit Test Condition
min.
max.
– 1.0
1.0
µA
2.0
VDD +
V
0 V ≤ VIN ≤ VDD
0.3
L-input level (except pins VIL1
SCLK, MCLK, DCLK)
– 0.3
0.8
V
H-input level (except pins VIH2
SCLK, MCLK, DCLK)
0.7 VDD
VDD
V
L-input level (except pins VIL2
SCLK, MCLK, DCLK)
0
0.3 VDD
V
H-output level
(except pins PZ1/PZ2)
VOH1
2.4
V
IO = 400 µA
H-output level
(pins PZ1/PZ2)
VOH2
VDD –
V
IO = 2 mA
Semiconductor Group
0.45
108
Electrical Characteristics
DC-Characteristics (cont’d)
VDD/VDDP = 5 V ± 5 %; VSSD/VSSA/VSSP = 0 V; TA = 0 to 70 ˚C
Parameter
Symbol
Limit Values
min.
Unit Test Condition
max.
L-output level
(except pin DU)
VOL1
0.45
V
IO = – 2 mA
L-output level
(pins DU, DD1))
VOL2
0.45
V
IO = – 7 mA
VDD supply current
IDDS1
0.9
mA
VDD = 5 V; DCL = ON
IDDS2
0.2
mA
DCL = OFF
standby (IOM-2 TE)
VDD supply current
operating (IOM-2 TE)
2)
Input capacitance
Output capacitance
VDD = 5 V
IDDO1
25
mA
IDDO2
25
mA
CI
CO
10
pF
15
pF
emergency ringing via
ALS (TGSR.TRL = 1)
handset mode
(ARCR.HOC = 010B)
1)
If voice channel swap (XOP_D) is enabled.
2)
Operating power dissipation is measured with all analog outputs open.
All analog inputs are set to VREF.
The digital input signal (pin DD) is set to an idle code.
For the emergency ringing mode, the tone generator is set to 400-Hz single tone (square).
In this mode the loudspeaker amplifier is set to – 3.5 dB (3.2 Vpp)
Note: Power dissipation values are target values.
AC-Characteristics
Inputs are driven to 2.4 V for a logical “1” and to 0.45 V for a logical “0”. Timing
measurements are made at 2.0 V for a logical “1” and 0.8 V for a logical “0”. The
AC-testing input/output waveforms are shown below.
Figure 38
Input/Output Waveforms for AC-Tests
Semiconductor Group
109
Electrical Characteristics
Analog Front End Input Characteristics
Parameter
Symbol
Limit Values
min.
typ.
Unit
Test Condition
kΩ
300 – 3400 Hz
max.
AMI-input impedance
ZAMI
AMI-input voltage swing
VAMI
19.3
mVpk
42 dB
AMI-gain
GAMI
42
dB
9.55 mV at 1 kHz
2
Ω
300 – 3400 Hz
15
Analog Front End Output Characteristics
AHO-output impedance
ZAHO
AHO-output voltage swing 1)
VAHO
3.2
Vpk
Load measured
from HOP to HON
AHO-output high voltage 1)
VAHOH
3.2
Vpk
input load – 1 mA
@ HOP/HON
AHO-output low voltage 1)
VAHOL
3.2
Vpk
input load + 1 mA
@ HOP/HON
Ω
300 – 3400 Hz
ALS-output impedance
2
ZALS
1)
VALS
3.2
Vpk
Load measured
from LSP to LSN
ALS-output high voltage 1)
VALSH
3.08
Vpk
input load – 60 mA
@ LSP/LSN
ALS-output low voltage 1)
VALSL
3.08
Vpk
input load + 60 mA
@ LSP/LSN
VREF output impedance
ZVREF
2
Ω
Load measured
from VREF to VSSA
VREF output voltage
VVREF
2.45
V
input load – 2 mA
@ VREF
ALS-output voltage swing
1)
2.35
The maximum output voltage swing corresponds to the maximum incoming PCM-code (± 127).
Semiconductor Group
110
Electrical Characteristics
Transmission Characteristics
VDD/VDDP = 5 V ± 5 %; VSSD/VSSA/VSSP = 0 V; TA = 0 to 70 ˚C
Parameter
Attenuation Distortion
@ 0 dBmO
Limit Values
min.
max.
0
– 0.25
– 0.25 0.25
– 0.25 0.45
– 0.25 0.9
0
Unit
Test Condition
dB
dB
dB
dB
dB
dB
< 200 Hz
200 – 300 Hz
300 – 2400 Hz
2400 – 3000 Hz
3000 – 3400 Hz
> 3400 Hz
receive:
4.6 kHz
8.0 kHz
transmit:
4.6 kHz
8.0 kHz
500 – 600 Hz
600 – 1000 Hz
1000 – 2600 Hz
2600 – 2800 Hz
0 to – 30 dB
– 40 dB
– 45 dB
3 to – 40 dB
– 40 to – 50 dB
– 50 to – 55 dB
receive (A-Law; Psoph.)
transmit (A-Law; Psoph.)
Reference: 0 dBmO
step accuracy
overall accuracy
Receive:
loudspeaker
earpiece
Transmit:
differential inputs
single ended input
Out-of-band signals
Group delay distortion
@ 0 dBmO 1)
Signal-to-total distortion
(method 2)
Gain tracking
(method 2)
@ – 10 dBmO
Idle-channel noise
Cross-talk
Programmable AFE gain
35
29
24
– 0.3
– 0.6
– 1.6
– 0.5
– 1.0
– 35
– 45
dB
dB
– 35
– 40
750
380
130
750
dB
dB
µs
µs
µs
µs
dB
dB
dB
dB
dB
dB
dBmO
dBmO
dB
dB
dB
0.3
0.6
1.6
– 75
– 66
– 66
0.5
1.0
Overall programming range
(with specified transmission – 21.5 11.5
characteristics)
– 21.5 2.5
0
0
1)
42
24
dB
dB
dB
dB
Delay measurements include delays through the A/D and D/A with all features filters FX, GX, FR
and GR disabled.
Semiconductor Group
111
Electrical Characteristics
IOM®-2 Bus Switching Characteristics
Figure 39
IOM®-2 Bus Timing Diagram
Parameter
Symbol
Limit Values
min.
DCL-clock period
1)
DCL-clock period
2)
tDCL
tDCL
DCL-duty cycle
FSC-period
FSC-setup time
FSC-hold time
DD-data-in setup time
DD-data-in hold time
DU-data-out delay
1)
2)
30
tFSC
tFSCs
tFSCh
tIDs
tIDh
tODd
typ.
max.
651
ns
244
ns
50
70
70
ns
40
ns
50
ns
50
ns
150
112
%
µs
125
768 kbit/s (IOM-2 TE-Mode); max. jitter of ± 160 ns once in FSC-period.
2048 kbit/s (IOM-2 Non-TE-Mode)
Semiconductor Group
Unit
ns
Electrical Characteristics
PCI-Switching Characteristics (IOM®-2 TE-Mode)
Figure 40
IOM®-2 Bus Timing Diagram (TE-Mode)
Parameter
Symbol
Limit Values
min.
PCI-data-out delay
PCI-data-in setup time
PCI-data-in hold time
Semiconductor Group
tPCld
tPCls
tPClh
113
Unit
max.
350
ns
50
ns
100
ns
Electrical Characteristics
SCI-Switching Characteristics
Figure 41
SCI-Switching Timing Diagram
Parameter
Symbol
Limit Values
min.
SCLK-frequency
Chip Select setup time
Chip Select hold time
SDR-setup time
SDR-hold time
SDX-data-out delay
SDX CS high to tristate
Semiconductor Group
fSCLK
tCSs
tCSh
tSDRs
tSDRh
tSDXd
tSDXt
114
Unit
max.
2048
kHz
0
ns
0
ns
50
ns
50
ns
150
ns
30
ns
Electrical Characteristics
SDI-Switching Characteristics
Figure 42
SDI-Switching Timing Diagram
Parameter
MCLK-frequency
DCLK-frequency
FSC-pulse width
FSC-hold time from DCLK low
FSC-delay time
DR-setup time
DR-hold time
DX-data-out delay (tFSCd < 0 ns)
DX-data-out delay (tFSCd ≥ 0 ns)
DX-data-out delay
Semiconductor Group
Symbol
fMCLK
fDCLK
tFSCw
tFSCh
tFSCd
tDRs
tDRh
tDXd1
tDXd1
tDXd
115
Limit Values
Unit
min.
max.
512
4096
kHz
64
4096
kHz
40
ns
30
ns
30
ns
50
ns
50
ns
80
ns
80 + tFSCd ns
80
ns
Package Outlines
6
Package Outlines
Plastic Package, P-DIP-28
GPD05037
(Dual-in-Line)
Sorts of Packing
Package outlines for tubes, trays etc. are contained in our
Data Book “Package Information”
Dimensions in mm
Semiconductor Group
116
Package Outlines
Plastic Package, P-DSO-28 (SMD)
GPS05123
(Dual Small Outline)
Sorts of Packing
Package outlines for tubes, trays etc. are contained in our
Data Book “Package Information”
SMD = Surface Mounted Device
Semiconductor Group
117
Dimensions in mm
Package Outlines
Plastic Package, P-LCC-28-1 (R) (SMD)
GPL05018
(Plastic Leaded Chip Carrier)
Sorts of Packing
Package outlines for tubes, trays etc. are contained in our
Data Book “Package Information”
SMD = Surface Mounted Device
Semiconductor Group
118
Dimensions in mm
Application Notes
Vakat
((120))
General Information
Table of Contents
Page
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
The Speakerphone Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Layout and Wiring Recommendation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
ARCOFI®-SP Telephone Board V1.0 SIPB 5132-SP . . . . . . . . . . . . . . . . . . . . . . . 181
Using the SIPB 5132-SP Telephone with the PSB 2163 . . . . . . . . . . . . . . . . . . . 209
ARCOFI®-SP Evaluation Board SIPB 5133-SP . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
ARCOFI®-SP Coefficients Software ARCOS-SP
and ARCOS-SP Plus SIPO 2163 V1.0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
ARCOFI®, EPIC®, IOM®, IPAT®, ISAC®, SICOFI® are registered trademarks of Siemens AG
Semiconductor Group
121
Vakat
((122))
Speakerphone Implementation
Introduction
The Siemens ARCOFI®-SP PSB 2163 is an Audio, Ringing, CODEC, Filter like the
ARCOFI® PSB 2160, but with additional integrated, enhanced speakerphone features,
which has been designed to save space, development costs, and time in any application
of digital terminal equipment (TE) featuring voice transmission, i.e. from the low-feature
telephone to the high-comfort telephone and the multifunctional terminal.
The present document has been conceived to help the digital TE-designer to take best
advantage of all the ARCOFI-SP-features.
This chapter consists of the application notes available on the ARCOFI-SP PSB 2163 as
well as a short description of the hardware and software tools developed, to give the
ARCOFI-SP user a good understanding of the component more quickly.
The present document does not replace the “ARCOFI-SP Technical Manual”; it is to be
used together with it.
Semiconductor Group
123
Vakat
((124))
The Speakerphone Implementation
Vakat
((126))
Speakerphone Implementation
Table of Contents
Page
1
1.1
1.2
1.3
1.4
1.5
1.6
General Aspects of Speakerphone Realizations . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Speakerphone Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Today’s Speakerphone Realizations . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Discussion of Practical Realizations of Half-Duplex Systems . . . . . . . . . .
Steady-State Problems and Transient Effects . . . . . . . . . . . . . . . . . . . . . .
Optimization of Half-Duplex Speakerphone Systems . . . . . . . . . . . . . . . .
129
129
130
131
132
134
135
2
2.1
2.2
2.3
2.4
2.5
2.6
2.7
Speakerphone with the PSB 2163 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Notation and Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Signal Flow Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Basics about the Speakerphone Algorithm . . . . . . . . . . . . . . . . . . . . . . . .
The Speech Detector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Speech Comparator for the Acoustic Echo . . . . . . . . . . . . . . . . . . . . .
The Speech Comparator for the Line Echo . . . . . . . . . . . . . . . . . . . . . . . .
137
137
138
139
140
142
147
152
3
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
Programming and Optimizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Speakerphone Test Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Programming Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Programming Amplifications in Speakerphone Mode . . . . . . . . . . . . . . . .
How to Determine GAE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Basic Rules for Optimizing the Speakerphone . . . . . . . . . . . . . . . . . . . . . .
Controlled Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Implementing a Volume Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Example Set of Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
154
154
155
156
157
159
161
163
164
Semiconductor Group
127
Speakerphone Implementation
About this Application Note
The intention of this application note is to give a thorough understanding of the
speakerphone implementation of the ARCOFI-SP. After introducing the general aspects
of speakerphones and the related problems (chapter 1) the ARCOFI-SP and all its
parameters are described (chapter 2). In chapter 3 practical aspects are discussed and
an example set of coefficients is given.
In the application note, the term "telephone" is used to describe the particular hardware,
the ARCOFI-SP is used for. Nevertheless, the ARCOFI-SP fits to a variety of
applications, like intercoms, videoconferencing systems etc.
Throughout this application note, registers or CRAM coefficients are written in bold and
capital letters (See “Notation and Coefficients” on page 138).
Some ARCOFI-SP parameters are related to the maximum PCM value. In the
application note A-law coding is assumed, therefore this value is written as 3.14 dBm0.
For µ-law coding 3.17 dBm0 would be correct.
Semiconductor Group
128
Speakerphone Implementation
1
General Aspects of Speakerphone Realizations
1.1
Introduction
Plain telephony suffers from the fact of being tied to a telephone set, and even mobile
telephony requires at least a handset to be held. To move freely for e.g. fetching some
documents needed during a discussion or when phoning while driving a car, the
speakerphone support is an important feature of telephone terminals.
High-feature telephones are mainly to be found as terminal devices in private exchanges
(PABX) and central offices (CO) to offer enhanced comfort compared to plain analog
terminals. In an Integrated Services Digital Network (ISDN), which opens access to a lot
of completely new services in the large list of terminal features, the speakerphone
support ranges on top and may be considered as a standard feature of comfort
terminals. But even high quality analog terminals take more and more advantage of
digital signal processing techniques and offer speakerphone support.
In contrast to using a handset which by its construction precludes feedback from
loudspeaker to microphone, the speakerphone feature opens such bypasses. It involves
much technical expense to cope with these various and time-varying couplings at a
reasonable performance level.
In general, today´s speakerphone realizations are based on a half-duplex transmission
mode which breaks the existing acoustic coupling between the receiver loudspeaker and
the speakerphone microphone by amplifying just one path (receive or transmit) while the
other one is attenuated.
Former realizations required a lot of discrete components to achieve a satisfying quality,
where "satisfying" is still subject to a rather personal sensation. Contemporary solutions
consist of an integrated circuit with still about 40 passive external components instead.
They reach an acceptable quality, but optimization is mainly empirical and tuning the
circuitry to the particular preferences of the user is impossible.
Controlled loudhearing, also called monitoring, is a valuable and very comfortable
feature, since it allows other people to follow a discussion running with a remote phone
partner. With the ARCOFI-SP, loudhearing can be regarded as a particular
speakerphone mode with reduced attenuation while using the handset microphone.
The digital realization of the speakerphone feature greatly simplifies the optimization and
tuning procedure by using the efficiency of a digital signal processor. The ARCOFI-SP
PSB 2163 incorporates the necessary computing capacity to realize a speakerphone of
superior performance without any external components. In the following sections the
typical coupling problems and technical solutions will be discussed, before in the second
part of this application note the concrete realization with the PSB 2163 is described in
detail.
Semiconductor Group
129
Speakerphone Implementation
1.2
Speakerphone Fundamentals
In a traditional telephone connection that uses handsets on both sides there is almost a
full duplex conversation (both subscribers can talk and listen at one and the same time),
and the gains required for the microphone and the earpiece are moderate.
As soon as people start to enjoy the handsfree telephony -moving free in the room, more
people listening and talking at the same time- the problems come up. The gain required
for amplifying microphone and loudspeaker signals is much higher and there are more
ways for the received signal to reach the microphone again than there are in a handset.
Unfortunately, any system at which an output signal can reach the input again starts to
oscillate if the loop gain increases 1 (and if a certain phase angle relation is fulfilled).
Especially the far-end echoes may be very annoying as they cause a delayed perception
of one’s own speech.
Figure 1
Coupling Problems Introduced by Speakerphones
The system shown in figure 1 should help to understand the different sources of echo
and delay. The speakerphone system on the left side is connected over a network to
either a handset or a speakerphone terminal at the far end. Amplifications are given in dB
with a positive sign, attenuation is marked with a negative sign. In a pure digital world
there is no 2/4-wire conversation, therefore the whole system is stable if the following
equation holds during conversation (not considering the delays):
GTN + GRN + AAN + GTF + GRF + AAF < 0 dB.
With reflections inside the network or with analog lines the situation becomes more
difficult and AHN, AHF must be considered as follows:
GTN + GRN + AAN + [AHN ||AHF ||(GTF + GRF + AAF)] < 0 dB.
Semiconductor Group
130
Speakerphone Implementation
Since there are a lot of possibilities for the signal to loop back in a speakerphone system,
efforts must be taken to keep the loop gain well below 0dB and to avoid regeneration.
Much technical expense is required to handle the different effects that occur in such a
system.
1.3
Today’s Speakerphone Realizations
Basically, it has to be distinguished between two different speakerphone realizations,
between the full-duplex and the half-duplex approach. A full-duplex system employs a
two-way transmission, whereas with a half-duplex system only a one-way transmission
is available at a time.
Full-Duplex
A full-duplex system is as comfortable as a round-table discussion, because the
interlocutors can talk and listen at the same time. The today's realizations like the
compander process or the echo cancellation cause enormous system costs and are very
critical, because in case of echo variations they react in a rather unpredictable manner.
Additionally, perceptible delays introduced by long calculation times due to the numeric
algorithm involved are annoying.
For this application note these remarks should be sufficient; more information can be
found in literature.
Half-Duplex
With a half-duplex system signal transmission is permitted only in one direction at any
time thus breaking the loop. The half-duplex process is stable under all circumstances,
and costs of such systems are remarkably lower than of full-duplex ones. A certain
discussion discipline, however, is required. Analog and digital realizations of the halfduplex system are available in form of integrated circuits.
The ARCOFI-SP contains a powerful digital implementation of an enhanced half-duplex
speakerphone system with an adjustable amount of switchable attenuation, therefore
the performance is somewhere between a full- and a half-duplex realization. The value
of the additional attenuation is programmable, and it is always switched into the nonactive path.
Semiconductor Group
131
Speakerphone Implementation
Figure 2
Half-Duplex Speakerphone with Switchable Attenuation AR and AT
1.4
Discussion of Practical Realizations of Half-Duplex Systems
To realize a high-quality half-duplex speakerphone, a sophisticated control mechanism
is necessary to decide where and when to use additional attenuation. This paragraph
explains the requirements to be fulfilled by such a control logic.
The switching between the transmit and receive direction must happen as discretely as
possible. This leads to the following requirements of the switching algorithm:
• Voice signals have to be recognized doubtless even if the background noise exceeds
the speech signals
• Ambient noise must not block the speakerphone switching process
• Speech direction switching must occur without any delay
• During a one-direction talk, the system must not switch prematurely
• Each speaker should be able to interrupt the other one
These specifications are partly contradictory, so that compromises have to be found.
Thus judging the speakerphone switching performance is rather subjective.
Semiconductor Group
132
Speakerphone Implementation
Transients
To get a satisfying switching performance, experience shows, that losses of the loop
should be kept as low as possible1). Ideally, a loss of about 10 - 20dB should be
sufficient and in reality this loss is not perceptible very well, so that a pseudo full-duplex
character can be achieved. In contrast, a switched attenuation larger than 40 dB gives
the feeling of speaking into a hole, like a dead line. In general, it is a fact that the lower
the volume of the speech comes out of the loudspeaker, the lower the inserted
attenuation may be. Of course, to the singing threshold there should always be a certain
reserve to avoid instability and annoying echoes.
For the switching control two speech detectors are necessary, one in the transmit path
and one in the receive branch. Both have to distinguish completely separately between
speech activity and (background) noise. The direction exhibiting a better S/N ratio should
normally be blocked and the path with a poorer S/N ratio should normally be activated.
Thus the path with a higher noise is normally transparent, and hence it will not be
necessary to have a large level difference between noise and speech. On the other
hand, in the path with a better S/N it is not difficult to discriminate between speech bursts
and noise edges.
A volume control must be arranged in such a way that, when turned down, it will reduce
the loss switched in both paths by the same amount. This guarantees that only the
minimum of loss is switched, which yields a comfortable switching feeling.
The sum of build-up and attack times should be very short (around 10 ms), the shorter
the better. The build-up time is the delay associated with the detection of a rising voice
signal, and the attack time (analogous release time) is the switching time required after
having detected speech. The sum of decay time (delay while detecting a decreasing
voice signal) and release time must be relatively long (150 ms)2). This prevents clipping
and does not delay mode switching in a way that it becomes noticeable.
1
A. Busla: "Fundamental Considerations in the Design of a Voice-Switched Speakerphone", The
Bell System Technical Journal; Vol. XXXIX, March 1969, Number 2
2
A. Busla: "Fundamental Considerations in the Design of a Voice-Switched Speakerphone", The
Bell System Technical Journal; Vol. XXXIX, March 1969, Number 2
Semiconductor Group
133
Speakerphone Implementation
1.5
Steady-State Problems and Transient Effects
Steady-state problems arise from the speakerphone system being stable either in
transmit mode or in receive mode. These problems are:
• Singing: occurs if the loss switched within the loop is too small (refer to figure 2)
• Transmit Blocking: the system remains in receive mode when it should have switched
to transmit mode; this can be due to large noise in the receive path or due to a small
AHN or AHF
• Receive Blocking: due to room noise or due to a strong coupling between loudspeaker
and microphone (refer to page 135) the circuit is prevented to switch from transmit to
receive mode
Transient effects occur during switching from one mode to another; these are:
• Initial Clipping: loss of the first syllables is caused by the operational time of the
speech detector and subsequent mode switching; it is influenced also by the
sensitivity of the speech detectors
• Final Clipping: the loss of the terminating parts of words is caused by the hangover
time
• The Starting Echo can be described as follows: speech signals are coming via the line
to the speakerphone system. The receive path is not attenuated, and there is a large
level enhancement due to the loudspeaker-microphone coupling. As a result, the level
at the transmit speech detector input is greater than that at the receive detector. Thus
when the non-speakerphone side starts speaking, the transmit speech detector can
recognize speech earlier than the receive detector. This may result in switching to the
transmit side (mode switching) and lead to some instability at speech onset.
• Similarly, the end echo can cause immediate mode switching. During a speech
interruption or pause in receive direction, the speakerphone system switches to
transmit mode. The reason is, that due to acoustic delays speech may still be effective
at the transmit speech detector when the receive speech detector has already
recognized "end of speech".
Additional problems arise from background noise and from speech collision during short
breaks in the flow of words or due to low-energy syllables. Also the varying coupling of
the speaker to the microphone may cause trouble at the transmit side.
Semiconductor Group
134
Speakerphone Implementation
1.6
Optimization of Half-Duplex Speakerphone Systems
A real speakerphone telephone set is a rather complex unit. Its performance is affected
both by electrical and acoustical couplings, the latter ones are set up via the surrounding
air and the telephone case itself. Arriving at a true optimum performance is only possible
if all trimming resources are exhausted.
The minimization of both, the acoustical feedbacks
and the electrical couplings is indispensable.
The minimization of both the acoustical feedbacks and the electrical couplings must be
done by acoustical as well as by electrical means. The gains made acoustically must not
be realized by programming the speakerphone hardware. It is not possible to optimize a
speakerphone system just by using only one method, acoustical or electrical. Starting
developing a speakerphone system is most effective in searching first for an economical
partitioning between both methods: only a sufficient acoustical optimization provides a
sound basis for completing the optimization electrically. Theoretically it is possible to
eliminate the near-end acoustical couplings almost completely by acoustical means, so
that only the hybrid couplings have to be controlled by loss switching.
Acoustical Optimizations
• Mechanical loop-back based on resonances of the telephone case (seismic coupling):
shock-proof mounting of the microphone and implementing some extra mounting
"towers" for obtaining a resonant-proof telephone case. Using soft materials for fixing
the microphone and loudspeaker to the telephone case.
• Acoustic loop-back via air inside the case: acoustic separation of critical case parts.
Insertion of additional walls inside the telephone case to block coupling via air.
• Acoustic loop-back due to the air external to the case: orienting the loudspeaker and
the microphone in different directions decouples effectively. An optimum is given, if
the lobe patterns of loudspeaker and microphone take an angle of 90˚.
• "Hollow Barrel" effect: placing the loudspeaker on the top towards the room and the
microphone near the bottom of the case. Placing the microphone into tubes, however,
generates unfavorable effects like reverberation.
• Maximum effectivity of the transmit amplification: the larger the opening on the
telephone case, the larger is the sound pressure at the microphone. Placing the
microphone in an angle of 45˚ to the horizontal towards the desk makes the
microphone receive the direct acoustical pressure and the indirect sound pressure
reflected at the desk. This results in an additional amplification of 4 - 6 dB.
• Acoustical feedback at the far-end talker side: this feedback is very small if a handset
is used. The attenuation via air due to the small amplifications in the earpiece and
handset microphone paths is about 50 dB, so that it can be neglected. If the far-end
Semiconductor Group
135
Speakerphone Implementation
side uses also a speakerphone system, an additional loss of 3dB is required. This loss
has always to be provided because it must be admitted that a speakerphone device is
connected at the far-end side.
Electrical Feedbacks
There are various electrical couplings which cause attenuated echoes with and without
delays. Referring to figure 1, couplings exist in the hybrid where the 4/2-wire conversion
is realized, and at the interfaces between analog and digital transmission lines.
External to the telephone case in pure digital connections electrical feedbacks do not
exist, but large delays can result from digital processing in the exchanges. Additional
delays may be introduced e.g. by an intercontinental link or by links via satellites. Though
these delays are very annoying, it is not possible to eliminate them. Moreover, in every
speakerphone they can generate enormous problems.
At the near-end hybrid (AHN) the attenuation of the echo varies from one link to another
in a range between 6 - 20dB. In the worst case it can decrease to just 6 dB. Due to the
relative small distance between the speakerphone system and the near-end hybrid, the
delay is negligible.
At the far-end hybrid the attenuation (AHF) of the generated echo can also reach 6 dB,
but via the analog line inserted, additional loss may be introduced. The still more
annoying effect is that the echo returns to the loudspeaker with some non-negligible
delay.
Semiconductor Group
136
Speakerphone Implementation
2
Speakerphone with the PSB 2163
2.1
Introduction
From the general description of speakerphone problems it can be seen that designing
and tuning a speakerphone system is a tricky task. A digital approach to solve the
problems not only makes the design independent of productional variations and of
environmental conditions, but also offers easy tuning to the desired response. Finally, a
single chip contains the complete circuitry, therefore no additional external components
are required. For the application of the speakerphone and controlled loudhearing
features all these necessary hard- and software can be found inside the ARCOFI-SP
PSB 2163.
A substantial feature of the ARCOFI-SP is to provide means for minimizing feedback in
electrical loops and to compensate for acoustical couplings. Therefore, the ARCOFI-SP
offers a lot of parameters which are set by software programming. No further external
components or trimming are required.
Note: Although the ARCOFI-SP allows the control of all kinds of feedback, it is strongly
recommended to have the majority of the acoustic couplings minimized by an
optimal design of the telephone case. The better the acoustic design, the better
the speakerphone will be.
The operational functions are realized with a signal processor where all necessary
parameters are parameterized. This technique offers a high level of flexibility and
reproducibility. An enormous advantage is the reduction of the optimization time. It is
possible to optimize the speakerphone system using just one link. The iterated
procedure of "building-up a line → testing → deactivating the link → opening the
telephone case → resoldering some RC-elements → closing the case" is reduced to a
comfortable computer session, focusing on the real work: the optimization of the
switching behavior.
As the signal flow graph of the ARCOFI-SP shows (see figure 3), the complete
operational algorithm is situated between the analog front end and the compression/
expansion logic. This offers the advantage that the speakerphone function is
independent of any country specific transmission characteristics. Thus telephone sets
can be optimized and adjusted to the particular geometrical and acoustic environment.
Furthermore, it is possible to connect different microphones to the speakerphone block,
so that controlled loudhearing may also be easily established.
Semiconductor Group
137
Speakerphone Implementation
The main features of the speakerphone signal processing are:
• Two separate attenuation stages, one for the transmit and one for the receive path,
controlled by an intelligent control logic, which processes information about the
current and the past speech activities;
• Two point speech detection; one speech detector for each path, with separate time
constants and separately programmable for signal and noise;
• Control of the acoustic echo performed by a fully programmable speech comparator;
• Control of the far-end echo performed by a fully programmable comparator for the line
echo;
• Background noise monitoring to eliminate continuous background noise from
decision.
The following chapters describe how the speakerphone implementation works and give
detailed hints about the values to be used for the different parameters. After discussion
of the block-diagram of the speakerphone, the speech detectors and the speech
comparators are described.
2.2
Notation and Coefficients
The ARCOFI-SP can be programmed with so called COP sequences ("Coefficient
Operation") and SOP sequences ("Status Operation"). Throughout this application note,
bold and uppercase abbreviations refer to CRAM coefficients to be programmed with
COP sequences (example: parameter ATT). Single bits or bit combinations to be set in
one of the registers are referred to as <Register>.<Bit> (example: GCR.SP). These bits
are always programmed with SOP sequences. For a description of all ARCOFI-SP
registers, please refer to the user’s manual.
The ARCOFI-SP offers a lot of CRAM coefficients which can be programmed. The hexvalue that corresponds to a certain physical value (e.g. B1H 49H representing 0dB
amplification in the GX stage) can be evaluated very comfortably with the ARCOS-SP
software or with the ARCOS-SP PLUS software (SIPO 2163) by pressing the right
mouse button. The ARCOS-SP PLUS software offers additionally hardware access to
program the ARCOFI-SP in different environments. The CRAM coefficients are not
completely documented in the user’s manual.
Semiconductor Group
138
Speakerphone Implementation
2.3
Signal Flow Graphs
Figure 3 shows the whole signal processor part of the ARCOFI-SP which interfaces to
the analog front end (AFE) on the left side.
Figure 3
Signal Flow Graph of the ARCOFI-SP Signal Processor (ASP)
Figure 4 gives an general idea of all signal processing blocks important for
understanding the speakerphone function. Some of the blocks like AGCX, AGCR
(automatic gain stages) and the attenuation stages GHX, GHR can be found directly in
the signal flow graph in figure 3. Others, like the two speech detectors SD and the
speech comparators SCAE, SCLE are located inside the block labeled "speakerphone
support" in figure 3.
Semiconductor Group
139
Speakerphone Implementation
Figure 4
Signal Flow Graph of the Whole Speakerphone Support Block
2.4
Basics about the Speakerphone Algorithm
This chapter explains the interaction of all the blocks seen in figure 4, and the three
different steady-state modes of the speakerphone. The inside of the speech detectors
(SD) and the speech comparators (SCAE, SCLE) including all programmable
parameters is explained in the following chapters.
Basically, we have to distinguish three steady-state modes with dynamic transitions
between them:
Table 1
Steady-State Modes when the Speakerphone is Active
Mode
Attenuation in GHX-stage
Attenuation in GHR-stage
TRANSMIT
yes, determined by ATT
no, 0dB
RECEIVE
no, 0dB
yes, determined by ATT
IDLE
yes, ATT / 2
yes, ATT / 2
The task of the attenuation control unit (see figure 4) is to decide which mode should be
effective. The information needed for decision is delivered by the two speech detectors
SD and the two speech comparators SCAE, SCLE. Assume the automatic gain control
stages AGCX and AGCR as switched off.
Semiconductor Group
140
Speakerphone Implementation
The speech detectors are able to distinguish between speech signals and any kind of
noise and inform the attenuation control unit about the presence or absence of speech
in the receive and the transmit path.
Assume the speakerphone is in receive mode, therefore outgoing digital signals are
attenuated in the GHX stage, incoming signals from the PCM interface are reproduced
by the loudspeaker with no attenuation inserted in the GHR stage. Of course the speech
detector in receive direction has recognized speech - otherwise the receive mode would
not be active. At the output of the microphone amplifier there is a signal mixture
consisting of coupled sound coming from the loudspeaker and from reflections in the
room. But also speech activity taking place at the near end in front of the speakerphone
will appear in the transmit path (which is still attenuated, if the ARCOFI-SP is in receive
mode). The task of the attenuation control unit is now to distinguish between unwanted
signals resulting from coupling, echoes etc. and true speech activity. True speech
activity has to make the ARCOFI-SP switch into transmit mode. For this purpose, first of
all the speech detector in transmit direction must have recognized any speech activity.
When both speech detectors have recognized speech activity, a decision must be made,
which mode is to use (receive or transmit). This is done by the speech comparators, in
this example by the speech comparator for the acoustic echo, SCAE. The speech
comparator for the acoustic echo "knows" exactly what signal level SX must be regarded
as an echo and which signal level is higher than an echo ever can be. If such a high
signal level appears, and this level is a certain (programmable) amount higher than the
signal level SR that is actually being received, the speech comparator will change state,
therefore indicating this fact to the attenuation control unit. As a result, mode switching
from receive mode to transmit mode will happen.
The same explanation applies to the mode switching from transmit mode to receive
mode, but in this case not the acoustic echo must be taken into account, but the line echo
resulting from delays and reflections in the telephone network is important. This echo is
handled by the speech comparator for the line echo, SCLE.
A mode switching can be forced simply by speaking louder than the remote partner. How
many "dBs" louder a person has to speak is programmable, but always the speech
comparators have to cover the worst case echo. That means for the application, that a
speakerphone system with a high acoustical coupling and with strong resonances will
show a reduced speakerphone performance, since for winning the channel one always
has to be louder than the signal caused by echoes and couplings. But for an appropriate
designed telephone plastics (refer to chapter 1.6), a very comfortable behavior regarding
mode switching is achieved.
If no speech activity is recognized by the ARCOFI-SP, the speakerphone goes into idle
mode. After both speech detectors have indicated "no speech", the ARCOFI-SP remains
in the previous mode for the period of time programmed by TW. After TW has passed,
the quantity of switchable attenuation (in dB) is changed to the amount of half the value
determined by ATT in both branches (see table below) with the speed given by DS
Semiconductor Group
141
Speakerphone Implementation
(decay speed in ms/dB). Therefore the transition into idle mode is very smooth and
hardly noticeable.
Starting from idle mode, depending on the first speech detector recognizing speech, the
device switches immediately into transmit or receive mode. The speed for this immediate
mode switching is determined by the parameter SW (switching time).
Table 2
Parameters for Switchable Attenuation / Speed of Transitions
Parameter
Typical Value
Meaning
ATT
10...40 dB
Amount of switchable attenuation that will always be
present in receive and/or transmit path (depending on
mode: transmit, receive, idle).
TW
144 ms
Wait time that passes before the transition from receive
or transmit mode into idle mode starts.
DS
99 ms/dB
Decay speed that determines how fast the switchable
attenuation changes when going into idle mode.
SW
0.6 ms/dB
Switching time for the transition out of idle mode into
transmit or receive mode or for the direct transition from
receive mode into transmit mode and vice versa.
Note, that the parameters explained above are not responsible for the decision, when to
change a mode. Any changes made for TW, DS, SW only affect the speed of the
transitions between the modes and are not responsible for example for the capability of
recognizing speech and switching the mode. This is handled by the speech detectors
and the speech comparators.
2.5
The Speech Detector
Both speech detectors have the same structure. For almost every application the
standard parameter set can be used, and there is usually no need for a change except
for the parameters described in the paragraph "processing of continuous tones".
Figure 5 shows the block diagram of the speech detector with all programmable
parameters. Basically, the speech detector makes use of the burst characteristic of
speech, that means, every fast change in signal amplitude compared to the average
signal level is recognized as speech. This is done by averaging the input signal with a
low pass filter (LP2) and comparing the output signal of this low pass with the input signal
itself.
Background Noise Monitor
Background noise monitoring is realized by the low pass LP2. Background noise can be
regarded as approximately stationary from its averaged power. Since only a difference
Semiconductor Group
142
Speakerphone Implementation
between the average signal level and the instant signal level leads to speech
recognition, the influence of noise on the decision is cancelled, even if the noise level
exceeds the voice level.
Figure 5
Block Diagram of the Speech Detector
Advanced Signal Processing
Before the signal is applied to the low pass LP2, it is preprocessed by a logarithmic
amplifier, a second low pass LP1, and a peak detector.
The logarithmic amplifier compands the signal logarithmically, therefore the speech
detector got a high dynamic range. With the parameter LIM a threshold is defined. Input
signal levels that are below the threshold determined by LIM are not processed by the
speech detector. Usually LIM is programmed for a threshold a few dB above the
system’s noise floor. Note, that LIM is related to the maximum PCM value, causing for
example a threshold of -60dBm0 if programmed to LIM= − 63.1 dBm0.
The main task of low pass filter LP1 is to filter the incoming signal in a way that main
spikes are eliminated. Due to the programmable time constant LP1 it is possible to
defuse as well high-energy syllables as noise edges.
The peak detector bridges the very short speech pauses during a monologue, so this
time constant has to be long. Furthermore, the speech bursts are stored, so that a sure
speech detection is guaranteed. But if no speech is recognized, the low pass LP2 must
be discharged very quickly to the averaged noise level. In addition, the noise edges are
Semiconductor Group
143
Speakerphone Implementation
to be smoothed. Therefore two time constants are necessary and have to be separately
programmable: PDS for speech and PDN for background noise signals. Thus "speech
mode" may be detected faster and kept longer than "no speech mode" so that smaller
breaks may not cause switching.
The low pass filter LP2 provides different time constants for noise (non-detected speech)
and speech. It determines the average of the noise reference level. In case of
background noise the level at the output of LP2 is approximately the level of the input.
Due to the offset OFF, the comparator remains in the initial state. In case of speech, the
difference of signal level between the offset branch and the LP2 branch at the
comparator increases and the comparator changes state. At speech bursts, the digital
signals arriving at the comparator via the offset branch change faster than those via the
LP2 branch so that the comparator changes its polarity. Hence two logical levels are
generated, one for speech and one for noise.
A small fade constant (LP2N) enables fast settling down the LP2 to the average noise
level after the end of speech recognition. It is recommended to choose a large rising
constant (LP2S) so that speech itself charges the LP2 very slowly. Generally, it is not
recommended to choose an infinite LP2S because then approaching the noise level is
disabled.
Figure 6
Example Showing the Principle of Speech Detection
The diagrams in figure 6 graphically describe the inputs and outputs of the speech
detector blocks. The result is not completely identical to the implementation in the
Semiconductor Group
144
Speakerphone Implementation
ARCOFI-SP, but helps to understand the function of the speech detectors. The first
diagram shows the input signal in the analog form because this form is better
understood. In the ARCOFI-SP this signal is delivered to the speech detector digitally.
The peak detector output is the envelope of the input signal, where short pauses are
bridged. The next diagram explains the task of the two branches with the offset OFF and
LP2, respectively of the background noise monitor. In the last diagram the result of the
determination of the complete speech detector can be seen.
Processing of Continuous Tones
Care has to be taken when continuous tones – e.g. signalling tones with a long
duration – or "music on hold" are to be processed by the speech detector. Any long
lasting signal that does not show the typical burst characteristic like speech will lead to
a "no speech" decision of the speech detector after the time constant LP2S has passed.
This is a necessary function for avoiding background noise to be recognized as speech.
With the help of the parameter LP2L the charging of the low pass LP2 can be limited to
a certain level, therefore a continuous signal that is greater than this limitation plus the
offset OFF will always be recognized as speech. This feature allows to detect any well
leveled signalling tone. Figure 7 shows the relationship between the input level and the
parameters.
Figure 7
Level Considerations for Processing Continuous Tones
Semiconductor Group
145
Speakerphone Implementation
The parameter LIM determines an input threshold for the speech detector and is related
to the maximum PCM value. The charging of LP2 is limited to a level of LP2L dBs above
the threshold LIM. A continuous input signal that must be recognized as speech has not
only to overcome line "a" in figure 7 but additionally the offset in the second branch of
the speech detector. Therefore any signal whose level is above the line "b" is high
enough always to be recognized as speech.
With the explanation given above, the required value for LP2L can be calculated if the
level of the continuous signal is known. Note that because of the averaging in LP2 a
calculation with peak values will always lead to a slight mismatch in the range of a few
dB.
Table 3
Parameters for the Speech Detectors
Parameter
Typical Value
Comment
LIM
− 54dB
Threshold, related to max. PCM value of + 3.14 dB;
signal values below this threshold are not processed.
LP1
4 ms
Time constant for low pass filter 1; suppression of
noise bursts and spikes; LP1 < PDN.
PDS
102 ms
Time constant for peak detector if speech has been
recognized; PDS > PDN for storing speech bursts and
bridging breaks during speech.
PDN
32 ms
Time constant for peak detector in case of noise.
LP2S
6.6 s
Time constant for low pass 2 if speech has been
recognized; a high value is necessary to avoid a quick
charging during speech.
LP2N
30 ms
Time constant for low pass 2 in case of noise; low time
constant allows quick adaption to changes of the
background noise level.
LP2L
25 dB
Limitation parameter; limits the charging of low pass 2.
OFF
4.5 dB
Offset in one branch of the speech detector; speech
bursts must be OFF dB above the average signal level
to be recognized.
Note that the ARCOFI-SP contains two speech detectors, one for the transmit direction,
one for the receive direction. Parameters for the speech detector in transmit direction are
usually marked with an additional letter "X", parameters for the speech detector in
receive direction are marked with an "R". Table 3 can be applied for both detectors.
Semiconductor Group
146
Speakerphone Implementation
2.6
The Speech Comparator for the Acoustic Echo
The speech comparator for the acoustic echo has the important task to decide, whether
a signal actually being received by the microphone is an echo or really a speech activity.
This is done by comparing the amplitude of the incoming signal with the amplitude of the
outgoing signal (See “Basics about the Speakerphone Algorithm” on page 140). If the
transmit signal is stronger than the signal being received at that moment, this will be
indicated to the speakerphone control logic which does a mode switching if also the
corresponding speech detector has recognized speech.
To understand how the comparator works it is assumed at first order, that mechanical
resonances can be neglected compared to the different kinds of acoustic echo. The
acoustic echo can be divided into two main parts. The first part of the signal reaches the
microphone directly from the loudspeaker, travelling inside or outside the telephone
housing (part 1). The second part results from reflections at the walls or boundaries like
the desk where the telephone is placed on (part 2). It takes about one millisecond for
signal part 1 to travel from loudspeaker to microphone (depending on the geometrical
arrangement). Reflected signals are not only weaker than the direct sound, they also
arrive with a certain delay. This situation is illustrated in figure 8.
Figure 8
Acoustic Echoes Resulting from Direct Coupling (1) or Reflections (2)
Semiconductor Group
147
Speakerphone Implementation
A signal burst fed into the loudspeaker amplifier appears after a short delay at the output
of the microphone amplifier (marked with a "1" in figure 8). Room echoes (marked with
"2") are significantly weaker and arrive later at the microphone. The resulting signal
envelope at the output of the microphone amplifier looks somehow like the bold line in
figure 8.
Therefore in first approximation the characteristic of the acoustical environment can be
described by a certain slope for the decrease of intensity of the echoes. But the
amplitude of the direct sound (signal part 1) arriving almost instantaneously at the
microphone depends on the amplification programmed for the ARCOFI-SP, the kind of
microphone and the loudspeaker used. As a consequence, the speech comparator for
the acoustic echo inside the ARCOFI-SP offers one parameter to compensate for these
amplifications (GAE, gain of the acoustic echo) and basically two additional parameters
to copy the acoustic properties of the room (GDAE, delta gain of the acoustic echo [dB]
and PDAE, peak decrement of the acoustic echo [dB/ms]). With these three parameters
the switching characteristic of the speech comparator is adjusted as shown in figure 9.
The characteristic is a picture of the acoustical phenomenons allowing a precise
differentiation between echo and true speech activity. Every signal whose level is above
the bold line in figure 9 will make the comparator to change state. If the signal level is
below the line, it must be regarded as an echo and no switching occurs.
Figure 9
Implementation of the Speech Comparator for the Acoustic Echo
Semiconductor Group
148
Speakerphone Implementation
Appropriate Values for the Comparator Parameters
From the explanations given above a set of values for GAE, GDAE, and PDAE can be
derived:
• After programming the desired amplifications for receive and transmit path, a
measurement of the terminal coupling loss in an anechoic room allows the calculation
of the required GAE (a detailed explanation follows in chapter 3.4); GAE then
compensates all amplifications in a way, that in case of direct coupling, the transmitted
and received signal shows the same amplitude at the dashed line in figure 10.
• The reverberation time of the acoustical environment determines roughly the slope of
the comparator characteristic; the reverberation time TN is defined as the period of
time in which the sound pressure level decays for 60dB after the sound source is
switched off. When assuming a reverberation time of about 0.5 seconds, a decrement
of PDAE = 8.5ms/dB is necessary (TN /60dB = 8.5ms/dB).
• Now GDAE is used to cover the sound that is directly coupled into the microphone,
this is the point marked with a black bullet in figure 9.
Figure 10
Signal Flow Graph of the Speech Comparator for the Acoustic Echo
Semiconductor Group
149
Speakerphone Implementation
It was assumed all the time, that mechanical resonances of the telephone housing can
be neglected. If this assumption does not hold, the adjustment of PDAE and GDAE must
be made in a way, that additionally the signal parts resulting from resonances are
covered by the comparator characteristic. In this case, other measurements or an
empirical approach must be used to find appropriate values. In general, a standard
parameter set is sufficient for most of the applications, but it is always advisable to adapt
GAE to the single telephone housing.
From the signal flow graph in figure 10 it becomes obvious that in fact there are two
programmable parameters each: GDNAE and GDSAE for the delta gain GDAE, PDNAE
and PDSAE for the peak decrement PDAE. Usually it is not necessary to take care of the
presence of two different parameters, and all explanations given in this chapter apply if
the same values are used for GDNAE and GDSAE as well as for PDNAE and PDSAE
the same values are programmed.
Note: To get a comparator characteristic as shown in figure 9, set GDSAE = GDNAE
and PDSAE = PDNAE. Don’t care for ETAE.
But of course there are applications where different values make sense, this is always
the case if an additional edge should be introduced in the comparator characteristic
shown in figure 9, for example, if very high mechanical resonances occur with a duration
that is short compared to the acoustic echoes. The letter "S" in GDSAE and PDSAE
stands for "speech" and means that these parameters are only effective, as long as
speech is recognized. Therefore it is possible to cover e.g. high mechanical resonances
with a high GDSAE and a relatively slow PDSAE but then switch to the other pair of
parameters, GDNAE and PDNAE, to get a fast decay ("N" stands for "noise"). Only in
this case with different values the echo time for the acoustic echo ETAE becomes
important. After the ARCOFI-SP has recognized "no speech" the period of time defined
by ETAE passes before the parameters for noise (GDNAE, PDNAE) are used.
Note: To get a comparator characteristic with an additional edge, use GDSAE/PDSAE
for the time during speech is recognized including the time ETAE afterwards. Use
GDNAE/PDNAE for the rest of the time where there is noise detected.
For almost every application it is not necessary to work with two sets of parameters, and
standard values can be used instead. But the ARCOFI-SP offers the flexibility to handle
even difficult coupling problems due to its two switchable coefficient sets for the speech
comparators.
Semiconductor Group
150
Speakerphone Implementation
Table 4
Parameters for the Speech Comparator for the Acoustic Echo
Parameter
Typical Value
Comment
GAE
– 5 … + 10 dB
Gain of the acoustic echo; used to achieve equal signal
levels at the comparator in case of direct coupling
sound, therefore representing the acoustic properties
of the telephone housing, microphone, and speaker;
can be calculated from a TCL measurement.
GDSAE
6 dB
Additional delta gain as reserve for GAE and as level
shift for the comparator characteristic (in case of
speech).
GDNAE
6 dB
Delta gain in case of noise.
PDSAE
8.5 ms/dB
Peak decrement of the acoustic echo in case of
speech; slope of the comparator characteristic used to
cover all echoes.
PDNAE
8.5 ms/dB
Peak decrement of the acoustic echo in case of noise.
ETAE
don’t care
Echo time for the acoustic echo; only effective in case
(see comment) of different parameters for GDSAE/PDSAE ("S")
GDNAE/PDNAE ("N"); determines the period of time
the "S" parameters remain active after speech
recognition has stopped before the "N" parameters are
used.
Semiconductor Group
151
Speakerphone Implementation
2.7
The Speech Comparator for the Line Echo
Having understood the task and principle of the comparator for the acoustic echo
(SCAE) the understanding of the speech comparator for the line echo (SCLE) is easy. As
shown in figure 2, the speakerphone has not only to deal with acoustic couplings but
also with echoes coming from the network. Basically, the intensity of echoes resulting
from reflections somewhere inside the telecommunication network or from the acoustic
connected to the far end, decreases exponentially. Therefore a comparator that has to
cover the line echo must show a characteristic similar to the one shown in figure 9 for
the acoustic echo.
The implementation of the SCLE is identical to the one of the SCAE and therefore all
explanations given in chapter 2.6 apply also to the comparator for the line echo. The
parameters to be used are different and result from the following considerations
(compare with the block diagram of the SCLE in figure 11).
Experience shows that in worst case the level of the echo coming from the line can be
about 10dB below the sending level. This level difference must be compensated with the
gain of the line echo (GLE) to get the same input conditions at the dashed line in
figure 11. Besides, the line echoes can show a significant long delay in the range of
some 100ms depending on the kind of connection used (consider satellite links). By
choosing a slow slope for the peak decrement PDLE and a high value for the delta gain
GDLE these echoes can be covered (see typical values in table 3).
Figure 11
Block Diagram of the Speech Comparator for the Line Echo
Semiconductor Group
152
Speakerphone Implementation
Table 5
Parameters for the Speech Comparator for the Line Echo
Parameter
Typical Value
Comment
GLE
− 10 dB
Gain of the line echo; used to achieve equal signal
levels at the comparator for covering the first echo; can
be measured as the level difference between outgoing
and received (echo-)signal at the network interface.
GDSLE
12 dB
Additional delta gain as reserve for GLE and as level
shift for the comparator characteristic (in case of
speech).
GDNLE
12 dB
Additional delta gain in case of noise.
PDSLE
21.3 ms/dB
Peak decrement of the line echo in case of speech;
slope of the comparator characteristic used to cover all
echoes.
PDNLE
21.3 ms/dB
Peak decrement of the line echo in case of noise
ETLE
don’t care
Echo time for the line echo; only effective in case of
(see comment) different parameters for GDSLE/PDSLE ("S")
GDNLE/PDNLE ("N"); determines the period of time
the "S" parameters remain active after speech
recognition has stopped before the "N" parameters are
used.
Semiconductor Group
153
Speakerphone Implementation
3
Programming and Optimizing
3.1
Speakerphone Test Function
At least during the optimization process it is highly recommended to make use of the
integrated speakerphone test mode of the ARCOFI-SP. It allows to observe the internal
state of the speakerphone control unit by the help of two LEDs connected to the piezopins of the ARCOFI-SP (see figure 12).
Figure 12
Test LEDs at the PZ-pins
Clipping and transient effects that are almost not audible can easily be detected with the
help of these LEDs. To activate the speakerphone test function, the bit TFCR.EPZST
and the bit TGSR.PM must be set. The test function has no influence on the functionality
of the ARCOFI-SP.
Table 6
Speakerphone States and Test LEDs
"TRANSMIT" LED
"IDLE" LED
Meaning
OFF
OFF
ARCOFI-SP is in receive mode
OFF
ON
Idle mode is active;
it was reached via the receive mode
ON
OFF
ARCOFI-SP is in transmit mode
ON
ON
Idle mode is active;
it was reached via the transmit mode
Semiconductor Group
154
Speakerphone Implementation
3.2
Programming Hints
COP Sequences
Table 7
COP Sequences and Relation to Speakerphone Parameters (bold)
COP
1. Byte
COP_A GAE
2. Byte
3. Byte
4. Byte
5. Byte
6. Byte
7. Byte
8. Byte
GLE
ATT
ETAE
ETLE
TW
DS
SW
COP_B GDSAE PDSAE GDNAE PDNAE GDSLE PDSLE GDNLE PDNLE
COP_C LIM
OFFX
OFFR
LP2LX
LP2LR
LP1X
LP1R
<xxx>
COP_D PDSX
PDNX
LP2SX
LP2NX
PDSR
PDNR
LP2SR
LP2NR
COP_E LGAX
COMX
AGX
TMHX
TMLX
NOISX
<xxx>
<xxx>
COP_F LGAR
COMR
AAR
AGR
TMHR
TMLR
NOISR
<xxx>
Except the ARCOFI-SP registers and COP_5 for the GX, GR amplification stages, all
bold faced parameters in table must be programmed to get the speakerphone work. The
corresponding hex values must be taken from the ARCOS-SP software. <XXX> are
don’t care bytes and should be set to 00H for future software compatibility.
Aborting COP Sequences
Any COP sequence can be interrupted without losing -with the particular sequence- the
already programmed parameters. If for example a COP_A sequence is aborted after the
first byte is successfully transmitted, the ARCOFI-SP will use the new value for GAE
transmitted within the first byte. This feature allows shortest programming times simply
by interrupting the sequence after the desired information has already been transmitted.
High Pass Filter
Especially for speakerphone mode it is necessary that the digital high-pass filters HPX
and HPR (compare with figure 3) are switched on (PFCR.DHPR=0, PFCR.DHPX=0),
for these digital high-pass filters are to remove the DC portion of the 16-bit word, which
e.g. could wrongly be regarded as background noise.
Generally these filters should be always switched on except for the very rare situations
when signalling at frequencies below 50Hz is used.
Semiconductor Group
155
Speakerphone Implementation
3.3
Programming Amplifications in Speakerphone Mode
It is important to notice, that as soon as the GCR.SP bit is switched on, the gain stages
LGAX and LGAR (located inside the automatic gain control stages) are activated
automatically. Figure 13 shows the gain stages responsible for amplification in transmit
or receive direction. The GHX and GHR stages for the switchable attenuation are not
shown.
Figure 13
Gain Stages in Speakerphone Mode
For measuring the over-all amplification in speakerphone mode it is necessary to
program the switchable attenuation ATT to 0dB (this is done by sending COP_A xx xx 00
xx xx xx xx xx, where "xx" are don’t care bytes).
Note: Throughout the ARCOFI-SP user’s manual absolute levels are given either in
dBm0 or in dBm. Levels given in dBm always refer to a 600 Ω load, therefore
0dBm is equivalent to 774.6 mVrms. The signal level in the digital domain is
usually expressed in dBm0. The maximum PCM signal level defined for A-law
coding is 3.14 dBm0 respectively 3.17 dBm0 for µ-law. The signal level in dBm
that is delivered from the D/A converter inside the ARCOFI-SP, if a 0 dBm0 PCM
signal is applied, is determined by the reference voltage of the converter (1.18 V)
causing a level difference of 20 log (1.18/0.775) [dBr] = 3.67dBr.
Level (analog)
Level (digital)
0.0 dBm
− 3.67 dBm0
3.67dBm
0.0 dBm0
Semiconductor Group
156
Speakerphone Implementation
3.4
How to Determine GAE
In the general description of the speech comparator for the acoustic echo the meaning
of the gain of the acoustic echo, GAE was introduced (see page 149). GAE should
always be adapted to the particular hardware (plastics, microphone, speaker) whereas
in first approximation for all the other parameters standard values can be used. This
chapter shows a simple measurement procedure to find an appropriate value for GAE.
Figure 14
Position of the Speech Comparators SCAE and SCLE
Figure 14 is important as it shows the position of the comparators related to the different
amplification stages. It can be seen that the comparator for the acoustic echo, SCAE,
senses the receive signal directly in front of the analog loudspeaker amplifier ALS and
the transmit signal between the GX and the LGAX stage. As explained in chapter 2.6 the
task of GAE is to adjust the signal level in a way, that in case of direct coupling the
amplitudes at the input of the SCAE are equal.
Semiconductor Group
157
Speakerphone Implementation
Measuring TCL and Calculating GAE
A measurement of the terminal coupling loss TCL1) is sufficient to calculate the required
value for GAE. A typical measurement procedure can be realized as follows:
• Programming of all amplifications (ATCR.MIC, GX, LGAX, ARCR.LSC, GR, LGAR).
Note, that for making LGAR and LGAX effective, the speakerphone must be enabled
(GCR.SP=1); to suppress any influence of the switchable attenuation ATT, set
ATT=0dB (COP_A xx xx 00 xx xx xx xx xx).
• Measuring the frequency dependent terminal coupling loss. This measurement has to
be done in an anechoic environment.
• Calculating GAE. Use the following equation with the programmed amplifications and
the measured TCL; in general it is sufficient to calculate with the weighted terminal
coupling loss TCLw, but if there are strong couplings at discrete frequencies, it can be
necessary to use a value between TCLw and the worse TCL values instead. Equation
for calculating GAE:
LGAX + TCL + LGAR + GR + GAE = 0 dB.
Example:
LGAX = 4.5 dB; LGAR = 3.0 dB; GR = 0dB; TCL = − 5 dB → GAE= − 2.5 dB.
Empirical Determination of GAE
If no anechoic room is available or for evaluation purposes also a simple test procedure
can lead to an appropriate value for GAE. After building up a speakerphone to handset
connection with negligeable line delays, GAE is varied and the acoustic effects on the
handset are evaluated. As a signal source either a human voice or synthetic speech
signals can be used (composite source signal, interrupted sinewave sweep, or switched
pink noise). If the telephone plastics does not exhibit an "extreme behavior", it should
always be possible to find a threshold value for GAE. If GAE is too low, the speech
comparator can not cover the acoustic echo and clipping effects occur. If GAE is
increased, the comparator will not change state due to echoes (compare with figure 9),
but the higher GAE, the more difficult it is to interrupt the signal from the speakerphone
side.
During tests, GAE should be changed only in small steps of e.g. 0.5dB because for
having the best speakerphone performance it is important to keep the value as low as
possible. This optimum value is usually in the range of − 5 dB … + 9 dB. The worse the
telephone housing (from the acoustical point of view), the higher the value for GAE must
be to cover the echo.
1
See: ETSI prl-ETS 300 245-3, "Technical characteristics of telephony terminals; Part3: PCM
A-law, loudspeaking and handsfree function", 4/93
Semiconductor Group
158
Speakerphone Implementation
The task is to find the minimum value for GAE for which the speakerphone works
properly.
3.5
Basic Rules for Optimizing the Speakerphone
General
• Always make use of the test function described in chapter 3.1.
• Start experiments with the standard coefficient set given in chapter 3.8.
• Make sure to have an appropriate value for GAE.
• Use a handset to speakerphone connection.
• Be sure to have programmed correct amplifications in the transmit and receive path
(see also the paragraph about volume control).
• The transmit and the receive direction can be checked separately, if the microphone
inputs or the loudspeaker outputs are disabled (ATCR.MIC=00H or ARCR.LSC=00H)
ATT
• The switchable attenuation ATT should be in the range of 20 dB … 30 dB; as
mentioned on page 133, the smaller ATT is, the better is the performance of the halfduplex system.
Speech Detectors
• The standard parameter set is not critical and shows best results under almost every
condition.
• With the ARCOFI-SP in receive or transmit mode, the test LED indicating "idle" during
a monologue must flash; this behavior is independent of the presence of background
noise.
• In case of high background noise which undergoes quick volume changes (e.g.
machines that are starting periodically) it could be interesting to allow the low pass
LP2 a quicker adaption to the average noise level (decrease LP2NX, but always keep
LP2SX < LP2NX); the disadvantage is, that speech detection becomes a bit more
insecure.
• If the system noise level – especially in receive direction – is quite high, thinking about
increasing the input threshold of the speech detector (LIMX) is necessary, but LIMX
must always be low enough to ensure a secure speech detection for the weakest
speech signal to be recognized.
Semiconductor Group
159
Speakerphone Implementation
Comparator for the Acoustic Echo, SCAE
• It’s most important to have a correct value for GAE; read chapter 3.4 for details
• SCAE is responsible for avoiding clipping effects: a monologue at the handset side
should always be transparent at the loudspeaker on the speakerphone side; the
"transmit"-LED must remain dark while the "idle"-LED is flashing; after the end of
speech at the handset side, the "transmit"-LED has to remain dark, otherwise
resonances or room echoes have caused the ARCOFI-SP to switch wrongly into
transmit mode
• Consider using a synthetic voice or tone signal (e.g. CSS or SPN from a test CD) since
these signals have a constant level
• Once having adjusted the comparator, it should be possible to interrupt a person
speaking at the far end simply by speaking loud(er) on the speakerphone side; the
lower the value for GAE is (respectively the better the telephone acoustics is), the
easier it is to interrupt the other side
• In case of clipping problems, it can be helpful to draw the comparator characteristic as
shown in figure 9 in a true scale on a piece of paper and think about possible
changes; if the reverberation time of the room is very long, increase PDAE; if the
telephone plastics shows strong resonances, increase GDAE; in worst case, a
second edge can be introduced in the comparator characteristic (see page 149)
Comparator for the Line Echo, SCLE
• The standard parameters are chosen under the assumption, that the worst case line
echo is at least 10 dB below the outgoing signal (GLE = – 10 dB) and that the line
delay for this high echo is shorter than 0.26 seconds (GDLE = 12 dB,
PDLE = 21.3 ms/dB, delay = GDLE×PDLE)
• Clipping effects due to a line echo are not to be expected with the parameters given
above, but if e.g. tests with a satellite link show unsatisfactory results, an increase of
GLE from e.g. − 10 dB to − 8 dB gives additional 2 dBs room for higher echoes; longer
delays have to be covered with higher GDLE and PDLE; again, this makes it more
difficult to interrupt a person talking at the speakerphone side
Semiconductor Group
160
Speakerphone Implementation
Typical Effects and the Corresponding Solutions
(See “Steady-State Problems and Transient Effects” on page 134; ↓ means
"decreasing the parameter, ↑ stands for "increasing")
Table 8
Typical Effects in Speakerphone Mode
Effect
Solution
Singing
ATT ↑
Transmit blocking
GAE ↓; GDSAE ↓; GDNAE ↓
Receive blocking
GLE ↓; GDSLE ↓; GDNLE ↓
Initial clipping
GDSLE ↓
Starting echo
GAE ↑; GDSAE ↑; GDNAE ↓
End echo
GLE ↑; GDSLE ↑; GDNLE ↑ ; PDSLE ↑; PDNLE ↑
Chopping effect
GAE ↑
3.6
Controlled Monitoring
"Controlled Monitoring" or "Loudhearing" can be regarded as a special speakerphone
mode with the difference, that the telephone user talks into the handset microphone and
the receive signal is reproduced by the handset earpiece and the loudspeaker. The
handsfree microphone is not active. Volume changes in the earpiece due to switchable
attenuation are not allowed.
The ARCOFI-SP supports Controlled Monitoring by using a slightly different attenuation
mechanism as in the speakerphone mode. To enable this feature, not only the GCR.SP
bit has to be set, but also the ARCR.CME bit (Controlled Monitoring Enable). The
Controlled Monitoring mode can be described as follows (compare the explanation with
figure 15):
• The speakerphone support is active and works as usual.
• The attenuation stage for receive direction GHR is fixed to a gain of 0 dB (this is in
contrast to the normal speakerphone mode and is necessary to have always a
constant volume at the handset output).
• The amplification in the analog loudspeaker amplifier ALS is determined by
ARCR.LSC as usual, but it can be switched to the fixed value of – 9.5 dB regardless
of the setting of ARCR.LSC.
• ALS is switched to – 9.5 dB automatically as soon as the speakerphone support of the
ARCOFI-SP has decided to use attenuation in the receive path; therefore the volume
of the signal coming out of the loudspeaker is reduced to a low volume when speech
activity takes place in the transmit path ("Controlled" Loudhearing).
Semiconductor Group
161
Speakerphone Implementation
Figure 15
Explanation of the Controlled Monitoring Mode
Hints for Optimizing
Generally, the controlled monitoring feature should be optimized after having found a
appropriate set of coefficients for the speakerphone. Then, after enabling the
ARCR.CME bit, only fine tuning is necessary. The system should be stable even if the
handset microphone is placed close to the loudspeaker. Due to the instantaneously
switching of the ALS, a reduction of DS and SW (decay speed, switching time) avoids
short-time instabilities. Note, that GAE must be adapted, if amplification in the
loudspeaker amplifier, the microphone amplifier or the GX stage is reprogrammed
(remember figure 14).
Semiconductor Group
162
Speakerphone Implementation
3.7
Implementing a Volume Control
The telephone user normally has the possibility to choose different volume settings when
using the speakerphone. The volume can be adjusted either manually by
reprogramming amplifications in the receive path or by means of the automatic gain
control stage. Since the automatic gain control stages are described in a separate
application note, this chapter deals with the first solution.
Table 9
Example for Eight Volume Steps
Vol. Step
8
7
6
5
4
3
2
1
total gain G
[dB]
G=
23
G-3=
20
G-6=
17
G-9=
14
G-12=
11
G-15=
8
G-18=
5
G-21=
2
ALS [dB]
11.5
8.5
5.5
2.5
2.5
2.5
2.5
2.5
GR [dB]
3.5
3.5
3.5
3.5
0.5
0
0
-0.5
LGAR [dB]
8
8
8
8
8
5.5
2.5
0
GAE [dB]
Y
Y-3
Y-6
Y-9
Y-9
Y-9
Y-9
Y-9
ATT [dB]
X
X-3
X-6
X-9
X-12
X-15
X-18
X-21
It’s important to keep some items in mind:
• In speakerphone mode, receive amplification can be changed with LGAR, GR, and
ARCR.LSC.
• For achieving best signal to noise performance, the maximum signal level should be
close to the max. PCM value; clipping must be avoided.
• It’s advisable, not to reduce the loop gain (which includes the switchable attenuation
ATT) when reducing the volume1); this can be achieved by reprogramming the
parameter ATT; as a consequence, the conversation gets the more transparent, the
lower the volume is.
• Every time, either ARCR.LSC, ATCR.MIC, or GX is changed, GAE must be adapted
too (applies only for speakerphone mode); see also figure 14.
Table 9 shows an example for eight volume steps with an over-all amplification in the
loudest volume step (step 8) of 23dB. For each volume step, the amplification is reduced
by 3dB. The value for GAE is shown as "Y" and must be reduced every time, the gain in
the ALS is reduced. The loop gain remains constant:
LGAR+GR+ARCR.LSC+ATCR.MIC+GX+LGAX+ATT = const.
1
A. Busla: "Fundamental Considerations in the Design of a Voice-Switched Speakerphone", The
Bell System Technical Journal; Vol. XXXIX, March 1969, Number 2, page 280
Semiconductor Group
163
Speakerphone Implementation
3.8
Example Set of Coefficients
The coefficients shown in table to should serve as a basis for every speakerphone
application. After starting with these values efforts must be taken to find an appropriate
value for GAE (see page 157). With the hints given in chapter 3.5 further adaption to the
particular requirements can easily be achieved.
Table 10
Standard Parameter Set for the Speech Detectors
Parameter
Value
Meaning
Coefficient
LP1X
4.0 ms
Low-pass 1
COP_C __ __ __ __ __ E1 __ __
PDNX
32.2 ms
Peak detector (noise) COP_D __ F4 __ __ __ __ __ __
PDSX
102.3 ms
dto. (speech)
COP_D 26 __ __ __ __ __ __ __
LP2LX
24.8 dB
Limitation for LP2
COP_C __ __ __ 42__ __ __ __
LP2NX
30.1 ms
Low pass 2 (noise)
COP_D __ __ __ 44 __ __ __ __
LP2SX
6.6 s
Low pass 2 (speech)
COP_D __ __ 20 __ __ __ __ __
OFFX
4.5 dB
Offset
COP_C __ 0C __ __ __ __ __ __
LIMX
– 54 dB
Limit (threshold)
COP_C 4_ __ __ __ __ __ __ __
LP1R
4.0 ms
Low-pass 1
COP_C __ __ __ __ __ __ E1 __
PDNR
32.2 ms
Peak detector (noise) COP_D __ __ __ __ __ F4 __ __
PDSR
102.3 ms
dto. (speech)
COP_D __ __ __ __ 26 __ __ __
LP2LR
24.8 dB
Limitation for LP2
COP_C __ __ __ __ 42__ __ __
LP2NR
30.1 ms
Low pass 2 (noise)
COP_D __ __ __ __ __ __ __ 44
LP2SR
6.6 s
Low pass 2 (speech)
COP_D __ __ __ __ __ __ 20 __
OFFR
4.5 dB
Offset
COP_C __ __ 0C __ __ __ __ __
LIMR
– 54 dB
Limit (threshold)
COP_C _4 __ __ __ __ __ __ __
Semiconductor Group
164
Speakerphone Implementation
Table 11
Standard Parameter Set for the Comparators
Parameter
Value
Meaning
Coefficient
GAE
5.3 dB
gain of acoustic echo
COP_A 0E __ __ __ __ __ __ __
ETAE
0.0 ms
echo time
COP_A __ __ __ 00 __ __ __ __
GDSAE
6.0 dB
delta gain (speech)
COP_B 20 __ __ __ __ __ __ __
PDSAE
8.5 ms/dB
peak decrement
COP_B __ 05 __ __ __ __ __ __
GDNAE
6.0 dB
delta gain (noise)
COP_B __ __ 20 __ __ __ __ __
PDNAE
8.5 ms/dB
peak dec. (noise)
COP_B __ __ __ 05 __ __ __ __
GLE
− 10.2 dB
gain of line echo
COP_A __ E5 __ __ __ __ __ __
ETLE
0.0 ms
echo time
COP_A __ __ __ __ 00 __ __ __
GDSLE
12.0 dB
delta gain (speech)
COP_B __ __ __ __ 40 __ __ __
PDSLE
21.3 ms/dB
peak decrement
COP_B __ __ __ __ __ 02 __ __
GDNLE
12.0 dB
delta gain (noise)
COP_B __ __ __ __ __ __ 40 __
PDNLE
21.3 ms/dB peak dec. (noise)
COP_B __ __ __ __ __ __ __ 02
Table 12
Standard Parameter Set - Other Pparameters
Parameter
Value
Meaning
Coefficient
ATT
28.2 dB
attenuation
COP_A __ __ 48 __ __ __ __ __
TW
144.0 ms
wait time
COP_A __ __ __ __ __ 09 __ __
DS
99.0 ms/dB decay speed
COP_A __ __ __ __ __ __ 25 __
SW
0.6 ms/dB
switching time
COP_A __ __ __ __ __ __ __ 64
LGAX
4.5 dB
gain adjustment
COP_E 13 __ __ __ __ __ __ __
LGAR
5.5 dB
gain adjustment
COP_F 12 __ __ __ __ __ __ __
Semiconductor Group
165
Speakerphone Implementation
Table 13
Example Register Setting for Speakerphone Mode
Register
Meaning
Coefficient
TFCR
Test function configuration
SOP_7 40
ARCR
AFE receive configuration
SOP_6 04
ATCR
AFE transmit configuration
SOP_5 70
TGSR
Tone generator switch
SOP_4 00
TGCR
Tone generator configuration
SOP_3 00
PFCR
Programmable filter configuration
SOP_2 00
DFICR
Data format and interface configuration
SOP_1 F1
GCR
General configuration
SOP_0 96
XCR
Extended configuration
SOP_A 00
If the SDI/SCI interface mode is used, in addition to the registers shown in table the
registers SDICR and TSCR have to be set first. Instead of sending eight particular SOP
sequences, a SOP_F can be used: SOP_F 40 04 70 00 00 00 F1 96.
All the bytes given in table 10 to 13 can be read from the ARCOS-SP Plus software1) or
from a previously saved file with the extension *.ARC. Table 14 shows the contents of
the ARC-file (not included COP_0 ... COP_9 because they are only necessary for the
tone generator and the AGC).
Table 14
Speakerphone Related Sequences to be Found in a ARC-File
W
W
W
W
W
W
W
0
0
0
0
0
0
0
COP_A
COP_B
COP_C
COP_D
COP_E
COP_F
SOP_F
0E
20
44
26
13
12
40
E5
05
0C
F4
84
84
04
1
ARCOS-SP software or ARCOS-SP PLUS software for the PSB 2163 (SIPO 2163)
Semiconductor Group
48
20
0C
20
10
7F
70
00
05
30
44
07
10
80
00
40
30
26
13
07
00
09
02
E1
F4
5F
13
00
25
40
E1
20
00
5F
F1
64
02
00
44
00
00
96
166
Layout and Wiring Recommendations
Vakat
((168))
Layout and Wiring Recommendations
Table of Contents
Page
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
2
Layout Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
3
3.1
3.2
3.3
Connecting the Analog Front End . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Outputs for Earpiece and Loudspeaker . . . . . . . . . . . . . . . . . . . . . . . . . . .
Differential Microphone Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Single Ended Input MI3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Semiconductor Group
169
175
175
176
179
Layout and Wiring Recommendations
1
Introduction
The ARCOFI-SP PSB 2163 is a high performance codec filter device with a state of the
art tone generator and an excellent speakerphone implementation. To obtain the full
performance of the device, some care in designing the analog circuitry and the printed
circuit board has to be taken. This application note gives some hints and wants to
provide understanding of the parameters that influence the performance.
With "performance", especially the following effects are meant:
• Signal to noise ratio, especially in transmit direction (analog to digital)
• Idle channel noise (the noise that is present when there is no signal applied)
• Spurious oscillations
• Sensitivity to any kind of interfering signals (noise on power supply, RF interference,
induced voltages etc.)
The ARCOFI-SP contains high performance A/D and D/A converters with more than
16bit resolution in order to allow the different signal processing steps without
degradation of the signals themselves. On one silicon the PSB 2163 integrates a high
gain, analog preamplifier, the converters as well as an digital signal processor. Besides
the ARCOFI-SP is used in an digital environment that typically causes noise on the
power supply and produces many kinds of interfering signals. For all these reasons,
careful grounding, decoupling, and shielding is the key to get best system performance.
Note: The circuits given in this application note are for general guidance and do not
claim to satisfy all user specific requirements. Especially for EMC reasons
additional components may be required.
Semiconductor Group
170
Layout and Wiring Recommendations
2
Layout Considerations
Since the ARCOFI-SP will be used with different kinds of printed circuit boards in
different applications and environments, it is not possible to show the "optimum layout".
Instead, this chapter explains the correlations that lead to the optimum layout for a given
application.
Power Supply Pins
The PSB 2163 has three ground pins and two pins for connecting the positive supply
voltage. Internally all the ground pins and all the power supply pins are connected. From
table 1 it can be seen, what parts of the ARCOFI-SP are supplied from what pins. It is
important, that each pair of pins given in one row of table 1 is decoupled with a pair of
capacitors in parallel. One capacitor has to be a 47 nF … 100 nF ceramic one, for the
second one a tantalum type with 1 µF … 10 µF is recommended.
Table 1
Power Supply Pins of the PSB 2163
Pin for VSS
Pin for VDD (+ 5 V) Supply for
1 (VSSD)
21 (VDD)
Digital signal processor and digital interface
15 (VSSP)
13 (VDDP)
Analog output amplifier AHO, ALS
20 (VSSA)
21 (VDD)
Analog preamplifiers and switches
Note, that pin 21 (VDD) is used for both, the supply of the DSP and the supply of the
analog part of the ARCOFI-SP except the power amplifiers AHO and ALS.
Ground Plane
A second aspect in designing a perfect PCB layout is a low impedance connection
between all points of the circuitry that have to be connected to ground. First of all, this
concerns the three ground pins of the ARCOFI-SP (pin 1, 15, 20). These three pins have
to be connected to each other directly underneath the IC package. This ensures not only
a low impedance ground for the PSB 2163 but also serves as a shield, if the connection
between the three ground pins is realized as a ground plane. Shielding is important since
it protects the high gain analog amplifiers and the converters against any capacitive
coupling. Hence, if multilayer PCBs are used, the ground plane has to be the top layer
and not somewhere in between.
Figure 1 shows one possibility to place the decoupling capacitors on a two layer board
if no components are allowed on the bottom of the PCB. If mounting on both sides is
permitted, one would place the capacitors between pin 1 and 21 on the bottom layer.
Semiconductor Group
171
Layout and Wiring Recommendations
Figure 1
Example Layout with Decoupling Capacitors and Ground Plane
Note the large ground area underneath the chip. It is a good approach to extend this
ground area also to the locations of all the passive components required for connecting
the acoustical transducers (microphones etc.). But any kind of ground loop must be
avoided since this would be an antenna for RF noise. Note also the position of the
decoupling capacitor for the reference voltage between pin 19 and 20 in figure 1. It
should be placed close to pin 19 and 20.
Semiconductor Group
172
Layout and Wiring Recommendations
Grounding of Microphones
Some transducers as electret microphones for example, always need a reference to
ground and therefore the question arises, which point of the circuit should serve as a
microphone ground.
As long as truly differential signal sources are used, this is not a problem. The high
common mode rejection ratio of the differential microphone inputs eliminates interfering
signals. But if an electret microphone is used, this microphone requires some biasing
and will be tied to ground with one pin (see chapter 3.2 for more examples about
connecting microphones). Figure 2 shows the typical arrangement for connecting
electret microphones. The microphone inputs MIN1/MIP1 and MIN2/MIP2 are
differential inputs with an input impedance of 15 kΩ or more. No external biasing is
required for these inputs. In the unbalanced configuration in figure 2, one of the inputs
is tied to the reference voltage (VREF pin). The reference voltage is a stable (2.4 V) and
very clean DC voltage; a current of up to 1.0 mA can be drawn from VREF. If the VREF pin
is used externally, it has to be blocked to the analog ground (VSSA) with a 100nF
capacitor (ceramic). If the VREF pin remains open, no blocking capacitor is required.
From figure 2 it becomes obvious, that any voltage difference introduced in the circle
marked with a dashed line, will be regarded as a wanted signal and therefore will be
amplified and converted by the ARCOFI-SP. In a real application, only the voltage
originating from the microphone itself has to be amplified. This leads to the requirement,
that the microphone ground must be connected directly to the pin VSSA (or to a ground
plane which incorporates VSSA).
If on a PCB the microphone is not placed directly next to the ARCOFI-SP, it is a good
solution to use a separate ground trace between the microphone and the analog ground
of the ARCOFI-SP. If the microphone is connected via a longer cable, the cable should
be shielded and again the screen is tied to the analog ground.
Summary
• Good decoupling between the pins given in one row of table 1
• Large ground plane underneath the chip connecting all ground pins
• Careful grounding of unsymmetrical signal sources
If these considerations receive attention, the full performance of the PSB 2163 is
available even in a very "noisy" environment.
Semiconductor Group
173
Layout and Wiring Recommendations
Figure 2
Ground Loop with Single Ended Signal Sources
Semiconductor Group
174
Layout and Wiring Recommendations
3
Connecting the Analog Front End
3.1
Outputs for Earpiece and Loudspeaker
The analog handset output amplifier AHO delivers a symmetrical signal at the pins HOP
and HON. Any load with an impedance higher than 200 Ω can be connected directly (see
figure 3a). However, if the load shows a strong capacitive behavior like a piezo-ceramic
earpiece, it is better to use series resistors as shown in figure 3b to avoid spurious
oscillations. This is also an typical example for a circuitry as it is used in a real
application. The resistors have to be placed close to the output pins.
Figure 3
The Analog Handset Output
It is possible to use the outputs HOP and HON as single ended outputs with reference to
ground; the load must be greater than 100 Ω and only half of the amplitude swing is
available.
Figure 4
The Analog Loudspeaker Output
Semiconductor Group
175
Layout and Wiring Recommendations
The difference between the earpiece output and the loudspeaker output is the driver
capability. The load between the pins LSP and LSN may be as low as 50 Ω in order to
drive a speaker. A dynamic speaker can be connected directly to these pins (see
figure 4a). An arrangement for an unsymmetrical connection with two speakers shows
figure 4b. The load can be 25 Ω but should be decoupled to avoid DC currents through
the speakers. With the register bits XCR.DLSP and XCR.DLSN each output pin can be
switched into a high impedance state therefore the speakers can be switched on and off
independently. The same applies for the earpiece output with HOP and HON. This
feature offers a variety of applications. For example, if one speaker in figure 4b is left
out, the output pin becomes a switchable line level output. For example, this output could
control an external speaker box.
3.2
Differential Microphone Inputs
The ARCOFI-SP offers five pins as microphone inputs. The two differential inputs
MIN1/MIP1 and MIN2/MIP2 are equivalent in terms of performance and circuitry. The
single ended input MI3 offers a slightly reduced performance due to its unsymmetrical
structure. All inputs can be used to interface directly with all kinds of microphones or
serve as a high level input.
Figure 5
Interfacing Symmetrical Microphones
A symmetric signal source that has no reference to ground can simply be connected to
one of the differential inputs. Basically, no additional components are required
(figure 5a). The microphone can be a dynamic, magnetic, or piezoelectric one. Usually
an arrangement similar to the one shown in figure 5b will be used in order to get well
defined impedances and a certain EMC protection. The component values depend on
the type of microphone used.
The microphone inputs are biased internally with VREF and require no external biasing.
They can be completely AC coupled. The input impedance is higher than 15 kΩ.
Semiconductor Group
176
Layout and Wiring Recommendations
Electret Microphones
Today, electret microphones are widely used in telecommunications devices. These
microphones usually contain an active amplifier or a FET to achieve a low output
impedance and therefore require some DC biasing.
The DC current to bias the electret microphone can be taken from the positive supply
voltage of the ARCOFI-SP (+ 5 V) or the reference voltage VREF can be used for this
purpose (2.4 V). The schematic in figure 6 depicts the first possibility.
Figure 6
Interfacing Electret Microphones with Bias from VDD
One input pin (MIN1) of the differential input is tied to VREF. This creates a single ended
configuration which must be AC coupled with C1 to the signal source, consisting of the
electret microphone. R1 and R2 are used to bias the microphone. The RC combination
R2/C2 filters out power supply ripple. R1 has the value recommended by the microphone
manufacturer and is usually in the range of 1 kΩ … 4 kΩ.
The dashed line in figure 6 illustrates how to connect a second microphone to the other
differential input of the ARCOFI-SP without spending the complete bias network again.
The reference voltage VREF has to be blocked with 100 nF to ground because any noise
at the pins MIN1 and MIN2 in the above configuration will be interpreted as a wanted
signal.
The second possibility to bias electret microphones is shown in figure 7. The DC current
is taken directly from the VREF pin of the ARCOFI-SP. No RC network to filter out power
noise is necessary because the reference voltage is clean and stable. Only in case of
high gains in the analog microphone amplifier (AMI more than 30 dB) the over-all
performance can be improved, if the reference voltage is filtered. For this purpose, a R/C
combination (1 kΩ/10 µF) between the VREF pin and the resistors R1 in figure 7 has to
be inserted. An application incorporating this RC element shows figure 8.
Semiconductor Group
177
Layout and Wiring Recommendations
Figure 7
Interfacing Electret Microphones with Bias from VREF
The solutions for connecting microphones shown in figure 6 and 7 contain a minimum
of components to explain the principle. A solution that comes closer to a real application
can be seen in figure 8. The microphone is biased from the VREF pin.
Figure 8
Example for a Real Application
With the help of R3/C3 a well defined input impedance is achieved. If R2 is inserted, a
high frequency cut-off can be realized. The 1 k/10 µF combination is used to filter the
reference voltage (improved idle channel noise with AMI > 30 dB).
Semiconductor Group
178
Layout and Wiring Recommendations
3.3
The Single Ended Input MI3
The single ended input MI3 behaves like one of the differential input pins, in case of one
input of the differential inputs is tied to VREF. Therefore the application examples in
figure 9 introduces no new ideas. An electret microphone can be biased from VDD
(figure 9a) or from the VREF pin (figure 9b). Any symmetric signal source with no
reference to ground should be connected between the MI3 pin and the VREF pin
(figure 9c). Often MI3 is used as a additional input e.g. for separate microphones with
their own preamplifiers. Such high-level signal sources could be connected to MI3 as
shown in figure 9d.
In general, if only two inputs are required, it is advisable to use MIN1/MIP1 and
MIN2/MIP2 for this purpose. They exhibit a slightly better performance because of their
differential nature. Unused inputs can be left open or tied to VREF.
Figure 9
Using the Single Ended Input MI3
Semiconductor Group
179
Vakat
((180))
ARCOFI®-SP Telephone Board V1.0 SIPB 5132-SP
- VAKATSEITE -
ARCOFI®-SP Telephone Board V1.0 SIPB 5132-SP
Contents
Page
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
2
Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
3
Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
4
4.0.1
4.0.2
4.0.3
4.0.4
4.0.5
4.0.6
4.0.7
4.0.8
4.0.9
Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Hook Switch Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DTMF-Generator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ARCOFI®-SP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Connector Pin-Outs of the Service Access Connector . . . . . . . . . . . . . . . . . . . . . . . .
Wiring Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
List of Replaceable Parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Floor Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
187
187
188
188
188
188
189
190
192
193
5
5.0.1
5.0.2
5.0.3
5.0.4
5.0.5
5.0.6
Operational Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Settings of Switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Settings of Jumpers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Programming the Peripheral Control Interface (PCI) . . . . . . . . . . . . . . . . . . . . . . . . .
TE-Mode Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
NT-S Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AC/DC-Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
194
194
195
195
196
196
197
6
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
7
7.0.1
7.0.2
7.0.3
7.0.4
Menu Software Track Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Track File HS_S02_I.TE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Track File HS_S02_S.TE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Track File SP_S02_S.TE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Track File HS_S0_S.NTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Semiconductor Group
183
199
199
202
204
207
SIPB 5132-SP
1
Introduction
The ARCOFI-SP Telephone Board SIPB 5132-SP V1.0 is part of a demonstration example of the
Siemens PC-User Board system SIPB to offer a complete solution for a digital telephone in the
ISDN-world of tomorrow. It represents the interface between the audio world – consisting of analog
transducers and amplifiers – and the digital world of PCM-coded transmission.
The kernel component of this board, the ARCOFI-SP PSB 2165, realizes all basic features like
dialing, ringing and voice transfer. Furthermore, thanks to a strictly applied digital technique, on the
same single chip full speakerphone feature without any additional external components are already
exhibited; all the required hardware and software has been implemented. Thus besides of working
on a single 5-V power supply, using the ARCOFI-SP PSB 2165 considerably eases the realization
of modern comfort telephones.
Semiconductor Group
184
SIPB 5132-SP
2
Features
● ARCOFI-SP PSB 2165, delivering the following features on-chip:
–
–
–
–
–
–
A/D-Conversion and Filtering
Programmable Analog Front End
Digital Speakerphone Support
DTMF-Generator
Ringing Generators
Comfortable Peripheral Control Interface
● DTMF-generation by the Dual-Tone Multi-Frequency Generator PSB 8593 for scanning the
12 + 1 keys keyboard
● Detection of hook switch changes by Monoflops 74 LS 221 in order to transmit a DTMF-signal
that normally starts procedures like activating the line
● IOM-2/SLD-interface to Audio Interface Module V2.0 and up SIPB 5130
3
Use
The ARCOFI-SP Telephone Board was originally conceived to be used at the terminal side of the
user board system (TE-configuration). Nevertheless it can also serve as the analogue front end in
the NT-S configuration due to the flexible concept of the SIPB-system. Thus the practical use of this
ARCOFI-SP Telephone Board SIPB 5132-SP is identical to that of the analogue telephone.
The ARCOFI-SP Telephone Board is always connected to the Audio Interface Module via the
IOM-2- or SLD-interface. Figure 1 shows how and in what combination it is connected via flat cable
to the mainboard.
Semiconductor Group
185
SIPB 5132-SP
Figure 1
Examples of TE- and NT-S-Configurations in the User Board Concept
Semiconductor Group
186
SIPB 5132-SP
4
Circuitry
4.1
Block Diagram
The block diagram of figure 2 below shows the main components of the ARCOFI-SP Telephone
Board. The whole circuitry can be divided into three blocks:
1. Hook Switch Logic
2. Key-Pad and DTMF-Generation
3. ARCOFI SP PSB 2165
Figure 2
Block Diagram of ARCOFI®-SP Telephone Board
Semiconductor Group
187
SIPB 5132-SP
4.1.1 Hook Switch Logic
The hook switch logic provides two features:
● signaling the state of the hook switch via pin 2 (SD) of the ARCOFI-SP and
● signaling a change of the hook switch position.
For a standard terminal application the first information is sufficient because the state is transparent
in the registers CIR1 in IOM-2-mode or in SSCR in SLD-mode respectively. In the Siemens ISDN
PC-userboard concept, however, the firmware needs an additional information about a change of
the hook switch position. This information is gained from a monoflop and an OR-gate. Every change
of the hook switch short-circuits a pair of pins of the DTMF-generator, so that a pair of frequencies
is transmitted to the DTMF-receiver of the Audio Interface Module SIPB 5130 ("relative detection"
of the hook switch) in order to activate/deactivate the line and power-up/power-down the
ARCOFI-SP.
4.1.2 DTMF-Generator
According to the key pressed, for each of the 12+1 push buttons on the single-contact keyboard the
Dual-Tone Multi-Frequency Generator IC1 (PSB 8593) generates pairs of frequencies. These are
specified by CCITT and are derived from a standard TV-crystal (3.58 MHz). The frequencies are
sent to the DTMF-receiver of the audio interface module.
4.1.3 ARCOFI®-SP
The Audio Ringing Codec Filter ARCOFI-SP (PSB 2165), the heart of the Telephone Board,
provides all necessary functions to convert analog signals into PCM-coded signals and vice versa.
In addition the highly sophisticated speakerphone feature is implemented into the same chip.
Setting gains and filter coefficients in receive and transmit direction, generating DTMF- and ringing
signals, amplifying the analog paths are all done by programming the device. A basic program is
shown in section 7. For more details on the ARCOFI-SP circuit and its programming structure refer
to the corresponding data sheet and description.
Transmission of PCM-coded signals is done either via lOM-2 bus or SLD-bus:
Using the IOM-2 bus two data lines are necessary: DU (Data Upstream) to send data from
ARCOFI-SP to layer 1, and DD (Data Downstream) to carry back data from layer 1 to the
ARCOFI-SP. While DU is physically equal to the SLD-line (pin 7 of ARCOFI-SP), DD is realized with
the line between pin 4 of the SAC-plug and pin 6 of ARCOFI-SP.
Using the SLD-bus the digital words are received/transmitted on the bidirectional SLD-line (Serial
Line Data) at pin 7 (SIP) of the ARCOFI-SP and led to pin 8 at the SAC-plug of the telephone board.
4.1.4 LEDs
Above the 12+1 keys keyboard of the ARCOFI-SP Telephone Board there are four LEDs to indicate
operational states. They are connected to the Peripheral Control Interface (PCI) of the ARCOFI-SP.
The left-most of these is to signal the state of the hook switch (SD). It lights up in off-hook position.
The remaining ones are for free use (see also section 5.3).
Semiconductor Group
188
SIPB 5132-SP
4.2
Connector Pin-Outs of the Service Access Connector
Figure 3
Pin-Outs of the Service Access Connector SAC
Pin
Function
1
Power supply + 5 V
2
Ground GND
3
n. c.
4
DD (Data Downstream)
5
Ringing
6
CLK (512 kHz) /DCL (1.536 MHz)
7
FSC (8 kHz)
8
SIP/DU (Data Upstream)
9
Reset
Semiconductor Group
189
SIPB 5132-SP
4.3
Wiring Diagram
Figure 4a
Wiring Diagram of the ARCOFI®-SP Telephone Board
Semiconductor Group
190
SIPB 5132-SP
Figure 4b
Wiring Diagram of the ARCOFI®-SP Telephone Board
Semiconductor Group
191
SIPB 5132-SP
4.4
List of Replaceable Parts
Component
Outline
Type / Value
IC1
IC2
IC3
IC4
(IC5
P-DIP-20
P-DIP-16
P-DIP-28
P-DIP-14
P-DIP-8
PSB 8593
74LS221
PSB2165
74LS32
NE5532)
T1 … T5
D1 … D4
TO92
V300
BSS89
HLMP1340
Q1
HC18
3.579545 MHz
R727
470 Ω
15 kΩ
56 kΩ
100 kΩ
75 Ω
4.7 kΩ
1.5 kΩ
7 × 2.7 kΩ
4 × 470 Ω
22 kΩ
R1
R2, R3, R5
R4
R 6, R 7
R 8, R 9
R10, R12
R11, R13
RP1
RP2
P1
10 µF
100 µF
1 µF
1 nF
10 µF
1 µF
100 nF
10 nF
2.2 nF
220 nF
3.3 nF
100 nF
C1 , C 2
C3
C4 , C 5
C6
C7
C8, C10
C9, C11
C12, C13
C14, C19
C15, C18
C16, C17
C20
Semiconductor Group
192
SIPB 5132-SP
4.5
Floor Plan
Figure 5
Floor Plan of the ARCOFI®-SP Telephone Board
Semiconductor Group
193
SIPB 5132-SP
5
Operational Information
All information about installing the Telephone Board, programming the ARCOFI-SP and the other
modules of the set-up in the various configurations is described in this section.
The ARCOFI-SP Telephone Board can be used in two different modes:
● Terminal Equipment TE-Mode
● Network Termination Simulator NT-S-Mode
Configuring the board for a particular mode of operation is done partly by setting switches, and
partly by software programming. For the peculiar Menu Software Track Files see section 7.
By setting jumpers accordingly, amplifiers can be inserted in the analog TX-transmission paths.
5.1
Settings of Switches
Two switches are used to configure the ARCOFI-SP Telephone Board for a desired mode of
operation. Settings are made before power is applied to the board.
In figure 6 the positions of switches and their particular functions are depicted.
Figure 6
Switches of the ARCOFI®-SP Telephone Board
Switches
SW1
SW2
SW3
Functions
selects the SLD- or IOM-2(TE)-mode in the ARCOFI-SP
selects the ARCOFI-SP IOM-2-address, either AD1 or AD2
Hook Switch
Semiconductor Group
194
SIPB 5132-SP
5.2
Settings of Jumpers
Two fields of four jumpers each are provided for inserting an OP-amplifier stage into each of the
analog transmit paths if the twin-amplifier IC 5 (NE 5532) is inserted. J1 connects the first amplifier
of IC 5 to the handset microphone path while J2 inserts the second one into the speakerphone path.
Figure 7
Settings of Jumpers J1 and J2
Left: The OP-Amp is Bypassed.
Right: The OP-Amp is Inserted into the Particular TX-Path.
Attention:
Before starting the initialization procedure with the menu software, please recheck the configuration
and switches on the modules to succeed in programming.
Verify that the ARCOFI-SP Telephone Board is always connected to that service access connector
that leads to the audio interface module.
5.3
Programming the Peripheral Control Interface (PCI)
According to the hardware configuration of the PCI the configuration register DFICR has to be
programmed by
DFICR = 8 × H
so that the Hook Switch Detection (SD) is an input while the other three ones (SA, SB, SC) are
outputs. The PCI is controlled via CIR1/CIX1 in IOM-2-mode or via SSCR/SSCX in SLD-mode.
Semiconductor Group
195
SIPB 5132-SP
5.4
TE-Mode Configuration
In the TE-configuration two serial interface modes – IOM-2 (TE) or SLD – can be used (refer to
section 4.1.3).
The configuration described supports both serial interfaces. Just the settings of the DlP-switches
select between IOM-2 (TE) – or SLD-mode.
Required hardware:
● a PC-Mainboard
SIPB 5000
● an Audio Interface Module V2.0 and up
SIPB 5130
IOM-2:
SLD:
DlP-switches 1 and 2 ON, others OFF
DlP-switch 1 ON, others OFF
● any layer-1 and link layer-2 modules
for example:
S-Access Module V2.1
IOM-2:
DlP-switch 4 ON, others OFF
SLD:
all DlP-switches OFF
SIPB 5100
For initializing this configuration the track files HS_S02_I.TE (lOM-2-mode) or HS_S02_S.TE (SLDmode) can be used. These two track files are listed in section 7.
Instead of using the S-Access Module SIPB 5100 any other layer-1 and link layer-2 modules can be
used too; if true please verify whether these modules support the serial interface mode.
5.5
NT-S Configuration
In the NT-S configuration only the SLD-mode is defined. The hardware consists of:
● a PC-Mainboard
SlPB 5000
● an Audio Interface Module V2.0 and up
SIPB 5130
● a Timing Module
SIPB 5310
● any layer-1 and link layer-2 modules
for example:
S-Access Module V2.1
SLD:
DlP-switch 1 ON, others OFF
SIPB 5100
The track file HS_S0_S.NTS sets up this configuration (see section 7).
Also in this mode, the S-Access Module SIPB 5100 can be replaced by any layer-1 and link layer2 modules. Please note, that in NT-S-configuration no IOM-2 (TE)-interface mode is defined.
Semiconductor Group
196
SIPB 5132-SP
5.6
AC/DC-Characteristics
The complete ARCOFI-SP Telephone Board works on a single + 5 V supply. A few layout hints may
be given which are important for a noiseless voice transmission (additional hints may be found in a
separate Application Note):
● Provide a large ground (GND) area underneath the ARCOFI-SP PSB 2165 where digital ground
VSSD and analog grounds VSSA and VSSP are connected.
● Insert ceramic blocking capacitors at the entry of the Telephone Board and right next to the
power pins of the ARCOFI-SP.
● Use sufficient large wire diameters and efficient isolation with the flat cable, which connects the
ARCOFI-SP Telephone Board with the Audio Interface Module.
● Separate digital blocks from analog circuitry.
Warning:
CMOS-lCs are very sensitive to electrostatic discharge. Never pull out the SAC-plug at the
telephone board before having switched off the power supply.
Semiconductor Group
197
SIPB 5132-SP
6
Glossary
ARCOFI-SP
CCITT
DC
DD
DTMF
DU
IC
IOM
ISDN
NT-S
PC
PCI
PCM
PIC
SAC
SIP
SIPB
SLD
TE
Semiconductor Group
Audio Ringing COdec FIlter with Speakerphone
Comité Consultatif International Télégraphe et Téléphone
Direct Current
Data Downstream
Dual-Tone Multi-Frequency
Data Upstream
Integrated Circuit
ISDN-Oriented Modular
Integrated Services Digital Network
Network Termination Simulator
Personal Computer
Peripheral Control Interface
Pulse Code Modulation
PC-Interface Connector
Service Access Connector
Serial Interface Port
Siemens ISDN PC-User Board System
Subscriber Line Digital
Terminal Equipment
198
SIPB 5132-SP
7
Menu Software Track Files
7.1
Track File HS_S02_I.TE
C **********************************************************************
C **********************************************************************
C*
C * Track File HS_S02_I.TE
C*
C * PROGRAMMING THE ARCOFI-SP VIA IOM2 in HANDSET MODE and additionally
C * ACTIVATING THE S0-INTERFACE
C*
C **********************************************************************
C ---------------------------------------------------------------------C ISAC-S in IOM2 mode
C ---------------------------------------------------------------------W /S02TE/ISAC_S/SERIAL/ADF2 80
W /S02TE/ISAC_S/SERIAL/SPCR 80
R /S02TE/ISAC_S/HDLC/STAR 4A
R /S02TE/ISAC_S/HDLC/STAR 4A
R /S02TE/ISAC_S/HDLC/STAR 4A
C ---------------------------------------------------------------------C Activation of the S0-Interface
C ---------------------------------------------------------------------W /S02TE/ISAC_S/SERIAL/CIX0 60
R /S02TE/ISAC_S/SERIAL/CIR0 30
R /S02TE/ISAC_S/SERIAL/CIR0 30
W /S02TE/ISAC_S/SERIAL/SPCR 05
R /S02TE/ISAC_S/SERIAL/SPCR 05
C ---------------------------------------------------------------------C IOM2-Identification of ARCOFI-SP
C ---------------------------------------------------------------------W /S02TE/ISAC_S/SERIAL/MOCR A0
R /S02TE/ISAC_S/SERIAL/MOSR 00
W /S02TE/ISAC_S/SERIAL/MOX1 A0
W /S02TE/ISAC_S/SERIAL/MOCR B0
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 00
R /S02TE/ISAC_S/SERIAL/MOSR A0
R /S02TE/ISAC_S/SERIAL/MOR1 A0
W /S02TE/ISAC_S/SERIAL/MOCR F0
R /S02TE/ISAC_S/SERIAL/MOSR 80
R /S02TE/ISAC_S/SERIAL/MOR1 82
R /S02TE/ISAC_S/SERIAL/MOSR 40
W /S02TE/ISAC_S/SERIAL/MOCR A0
R /S02TE/ISAC_S/SERIAL/MOSR 00
Semiconductor Group
199
SIPB 5132-SP
C ---------------------------------------------------------------------C Programming the ARCOFI-SP in HANDSET mode
C ---------------------------------------------------------------------R /S02TE/ISAC_S/SERIAL/MOSR 00
W /S02TE/ISAC_S/SERIAL/MOX1 A1
W /S02TE/ISAC_S/SERIAL/MOCR B0
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 1F
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 00
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 30
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 50
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 00
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 80
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 20
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 82
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 1B
R /S02TE/ISAC_S/SERIAL/MOSR 20
C ---------------------------------------------------------------------C COP_7: GZL=-00, GZH=-16dB
C ---------------------------------------------------------------------W /S02TE/ISAC_S/SERIAL/MOX1 27
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 97
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 12
R /S02TE/ISAC_S/SERIAL/MOSR 20
C ---------------------------------------------------------------------C Powering up the ARCOFI-SP
C ---------------------------------------------------------------------W /S02TE/ISAC_S/SERIAL/MOX1 10
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 1E
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOCR A0
R /S02TE/ISAC_S/SERIAL/MOSR 00
W /S02TE/ISAC_S/SERIAL/CIX1 20
Semiconductor Group
200
SIPB 5132-SP
C ---------------------------------------------------------------------C Read Out CR: 00, 30, 50, 00
C
80, 20, 82, 1E
C ---------------------------------------------------------------------W /S02TE/ISAC_S/SERIAL/MOX1 A1
W /S02TE/ISAC_S/SERIAL/MOCR B0
R /S02TE/ISAC_S/SERIAL/MOSR 20
W /S02TE/ISAC_S/SERIAL/MOX1 9F
R /S02TE/ISAC_S/SERIAL/MOSR A0
R /S02TE/ISAC_S/SERIAL/MOR1 A1
W /S02TE/ISAC_S/SERIAL/MOCR F0
R /S02TE/ISAC_S/SERIAL/MOSR 80
R /S02TE/ISAC_S/SERIAL/MOR1 00
R /S02TE/ISAC_S/SERIAL/MOSR 80
R /S02TE/ISAC_S/SERIAL/MOR1 30
R /S02TE/ISAC_S/SERIAL/MOSR 80
R /S02TE/ISAC_S/SERIAL/MOR1 50
R /S02TE/ISAC_S/SERIAL/MOSR 80
R /S02TE/ISAC_S/SERIAL/MOR1 00
R /S02TE/ISAC_S/SERIAL/MOSR 80
R /S02TE/ISAC_S/SERIAL/MOR1 80
R /S02TE/ISAC_S/SERIAL/MOSR 80
R /S02TE/ISAC_S/SERIAL/MOR1 20
R /S02TE/ISAC_S/SERIAL/MOSR 80
R /S02TE/ISAC_S/SERIAL/MOR1 82
R /S02TE/ISAC_S/SERIAL/MOSR 80
R /S02TE/ISAC_S/SERIAL/MOR1 1E
R /S02TE/ISAC_S/SERIAL/MOSR 40
W /S02TE/ISAC_S/SERIAL/MOCR A0
R /S02TE/ISAC_S/SERIAL/MOSR 00
C **********************************************************************
C*
C*
End of Track File
C*
C **********************************************************************
C **********************************************************************
Semiconductor Group
201
SIPB 5132-SP
7.2
Track File HS_S02_S.TE
C **********************************************************************
C **********************************************************************
C*
C * Track File HS_S02_S.TE
C*
C * PROGRAMMING THE ARCOFI-SP VIA SLD in HANDSET MODE and
C * additionally ACTIVATING THE S0 - INTERFACE
C*
C **********************************************************************
C **********************************************************************
C
C ---------------------------------------------------------------------C Activation of the S0-interface
C ---------------------------------------------------------------------W /S02TE/ISAC_S/SERIAL/CIX0 60
R /S02TE/ISAC_S/SERIAL/CIR0 30
R /S02TE/ISAC_S/SERIAL/CIR0 30
W /S02TE/ISAC_S/SERIAL/SPCR 45
R /S02TE/ISAC_S/SERIAL/SPCR 45
C ---------------------------------------------------------------------C Programming the ARCOFI-SP in HANDSET mode
C ---------------------------------------------------------------------B 1F
B 00
B 30
B 50
B 00
B 80
B E0
B 82
B 1A
X /S02TE/ISAC_S/BUS/CONTR
C ---------------------------------------------------------------------C COP_5: GX= 00dB
C ---------------------------------------------------------------------D
B 25
B A0
B 01
X /S02TE/ISAC_S/BUS/CONTR
Semiconductor Group
202
SIPB 5132-SP
C ---------------------------------------------------------------------C COP_6: GRL= 00dB, GRH= 00dB
C ---------------------------------------------------------------------D
B 26
B A0
B 01
B A0
B 01
X /S02TE/ISAC_S/BUS/CONTR
C ---------------------------------------------------------------------C COP_7: GZL=–54dB, GZH=–16dB
C ---------------------------------------------------------------------D
B 27
B 97
B 12
X /S02TE/ISAC_S/BUS/CONTR
C ---------------------------------------------------------------------C Powering up the ARCOFI-SP
C ---------------------------------------------------------------------D
B 10
B 1E
X /S02TE/ISAC_S/BUS/CONTR
W /S02TE/ISAC_S/SERIAL/SSCX 20
C ---------------------------------------------------------------------C Read Out CR: 00, 30, 50, 00
C
80, E0, 82, 1E
C ---------------------------------------------------------------------R /S02TE/ISAC_S/BUS/CONTR 9F 08
A 0030 5000 80E0 821E
C
C **********************************************************************
C **********************************************************************
C*
C*
End of Track File
C*
C **********************************************************************
C **********************************************************************
Semiconductor Group
203
SIPB 5132-SP
7.3
Track File SP_S02_S.TE
C **********************************************************************
C **********************************************************************
C*
C * Track File SP_S02_S.TE
C*
C * PROGRAMMING THE ARCOFI-SP VIA SLD in SPEAKERPHONE MODE and
C * additionally ACTIVATING THE S0 - INTERFACE
C*
C **********************************************************************
C **********************************************************************
C
C ---------------------------------------------------------------------C Activation of the S0-interface
C ---------------------------------------------------------------------W /S02TE/ISAC_S/SERIAL/CIX0 60
R /S02TE/ISAC_S/SERIAL/CIR0 30
R /S02TE/ISAC_S/SERIAL/CIR0 30
W /S02TE/ISAC_S/SERIAL/SPCR 45
R /S02TE/ISAC_S/SERIAL/SPCR 45
C ---------------------------------------------------------------------C Programming the ARCOFI-SP in SPEAKERPHONE mode
C ---------------------------------------------------------------------D
B 1F
B 00
B 02
B 62
B 00
B 80
B 00
B 83
B BA
X /S02TE/ISAC_S/BUS/CONTR
Semiconductor Group
204
SIPB 5132-SP
C ---------------------------------------------------------------------C COP_B: SDX
C ---------------------------------------------------------------------D
B 2B
B E1
B 34
B D3
B 75
B 26
B F2
B 20
B 44
X /S02TE/ISAC_S/BUS/CONTR
C ---------------------------------------------------------------------C COP_C: SDR
C ---------------------------------------------------------------------D
B 2C
B E1
B 34B D3
B 00
B 26
B F2
B 20
B 44
C ---------------------------------------------------------------------C COP_D: ATTENUATION CONTROL
C ---------------------------------------------------------------------D
B 2D
B 35
B 2B
B 04
B FF
X /S02TE/ISAC_S/BUS/CONTR
W /S02TE/ISAC_S/SERIAL/SSCX 10
C ---------------------------------------------------------------------C COP_E: AGCX
C ---------------------------------------------------------------------D
B 2E
B 14
B 13
B 15
B 80
Semiconductor Group
205
SIPB 5132-SP
C ---------------------------------------------------------------------C COP_F: LGA
C ---------------------------------------------------------------------D
B 2F
B 90
B 00
B 00
B 00
X /S02TE/ISAC_S/BUS/CONTR
C ---------------------------------------------------------------------C Powering up the ARCOFI-SP
C ---------------------------------------------------------------------D
B 10
B BE
X /S02TE/ISAC_S/BUS/CONTR
C ---------------------------------------------------------------------C Read Out CR: 00, 02, 62, 00
C
80, 00, 83, BE
C ---------------------------------------------------------------------R /S02TE/ISAC_S/BUS/CONTR 9F 08
A 0002 6200 8000 83BE
C
C **********************************************************************
C **********************************************************************
C*
C*
End of Track File
C*
C **********************************************************************
C **********************************************************************
Semiconductor Group
206
SIPB 5132-SP
7.4
Track File HS_S0_S.NTS
C **********************************************************************
C **********************************************************************
C*
C * Track File HS_S0_S.NTS
C*
C * PROGRAMMING THE ARCOFI-SP VIA SLD in HANDSET MODE and
C * additionally ACTIVATING THE S0 - INTERFACE
C*
C **********************************************************************
C **********************************************************************
C
C ---------------------------------------------------------------------C Activation of the S0-interface
C ---------------------------------------------------------------------W /S02NTS/ISAC_S/SERIAL/CIX0 60
R /S02NTS/ISAC_S/SERIAL/CIR0 30
R /S02NTS/ISAC_S/SERIAL/CIR0 30
W /S02NTS/ISAC_S/SERIAL/SPCR 45
R /S02NTS/ISAC_S/SERIAL/SPCR 45
C ---------------------------------------------------------------------C Programming the ARCOFI-SP in HANDSET mode
C ---------------------------------------------------------------------D
B 1F
B 00
B 30
B 50
B 00
B 80
B E0
B 82
B 1A
X /S02NTS/ISAC_S/BUS/CONTR
C ---------------------------------------------------------------------C COP_5: GX= 00dB
C ---------------------------------------------------------------------B A0
B 01
X /S02NTS/ISAC_S/BUS/CONTR
Semiconductor Group
207
SIPB 5132-SP
C ---------------------------------------------------------------------C COP_6: GRL= 00dB, GRH= 00dB
C ---------------------------------------------------------------------D
B 26
B A0
B 01
B A0
B 01
X /S02NTS/ISAC_S/BUS/CONTR
C ---------------------------------------------------------------------C COP_7: GZL=-54dB, GZH=-16dB
C ---------------------------------------------------------------------D
B 27
B 97
B 12
X /S02NTS/ISAC_S/BUS/CONTR
C ---------------------------------------------------------------------C Powering up the ARCOFI-SP
C ---------------------------------------------------------------------D
B 10
B 1E
B /S02NTS/ISAC_S/BUS/CONTR
B /S02TE/ISAC_S/SERIAL/SSCX 20
C ---------------------------------------------------------------------C Read Out CR: 00, 30, 50, 00
C
80, E0, 82, 1E
C ---------------------------------------------------------------------R /S02NTS/ISAC_S/BUS/CONTR 9F 08
A 0030 5000 80E0 821E
C
C **********************************************************************
C **********************************************************************
C*
C*
End of Track File
C*
C **********************************************************************
C **********************************************************************
Semiconductor Group
208
Using the SIPB 5132-SP Telephone with the PSB 2163
Vakat
((210))
PSB 2163
Table of Contents
Page
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
2
Using the SIPB 5132-SP Telephone with the
PSB 2163 in IOM-2 TE Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
3
Using the SIPB 5132-SP Telephone with the PSB 2163
in IOM-2 NON-TE Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Semiconductor Group
211
PSB 2163
1
Introduction
About this Application Note
The SIPB 5132-SP telephone1) is a telephone for evaluation purposes, showing all
features of the ARCOFI-SP PSB 2165 and PSB 2163. This telephone is equipped with
the ARCOFI-SP PSB 2165 and the explanations given in the documentation of the
SIPB 5132-SP telephone are related to the PSB 2165.
With the PSB 2163 a pin compatible device is available, which offers a completely new
speakerphone support. The speakerphone implementation of the PSB 2163 has
superior performance and enables the telephone user to have very comfortable
speakerphone to speakerphone discussions.
The SIPB 5132-SP telephone can be used with both ICs, with the PSB 2165 (with which
it comes along) and with the PSB 2163 (which can be build into the telephone).
This document describes how to replace the PSB 2165 with the PSB 2163. This is not
an instruction that stipulates each step that must be taken, this is rather an explanation
of the things to be taken into account when replacing the PSB 2165 with the PSB 2163.
Differences between PSB 2165 and PSB 2163
Concerning the SIPB 5132-SP Telephone
For a detailed description of the two ICs, please refer to the user’s manual of the
PSB 2165 and the user’s manual of the PSB 2163. This paragraph only gives
information about differences that can affect the external hardware circuitry. Basically,
both ARCOFI-SPs are fully pin compatible and replacing is possible without any
difficulties.
The interface modes that can be used with the PSB 2165 are the IOM-2 TE interface
(1.536 MHz), and the SLD interface. Together with the PSB 2163 and the
SIPB 5132-SP telephone, the IOM-2 TE interface can be used as well as the IOM-2
NON-TE interface (4.096 MHz).
There is a difference concerning the function of the PCI-interface pins (pin SA, SB, SC,
SD) between the two ARCOF-SPs; see table 1 for details.
1
Ordering Code Q67100-H6299
Semiconductor Group
212
PSB 2163
A slight change concerns the names of the microphone inputs. Although the functionality
is the same, the names are different as table 2 shows.
Table 1
Use of the Pins SA, SB, SC, SD
Device
Interface
Meaning of Pins SA, SB, SC, SD
2165
IOM-2 TE
PCI-interface;
SA, SB can also be used to connect test LEDs
2165
SLD
PCI-interface;
SA, SB can also be used to connect test LEDs
2163
IOM-2 TE
PCI-interface;
(test LEDs can be connected to the PZ1, PZ2 pin)
2163
IOM-2 NON-TE
SB, SC, SD used for slot select; SA unused (VSS );
(test LEDs can be connected to the PZ1, PZ2 pin)
Table 2
Microphone Inputs
Pin
PSB 2165
PSB 2163
Used for (SIPB 5132-SP)
8
XINP
MIP1
Not connected
9
XINN
MIN1
Not connected
10
MIN
MIN2
Handset microphone
11
MIP
MIP2
12
FHM
(single-ended)
MI3
(single-ended)
Semiconductor Group
Hands-free microphone
213
PSB 2163
2
Using the SIPB 5132-SP Telephone with the PSB 2163 in IOM-2 TE Mode
The basic steps to be performed are quite simple:
•
•
•
•
Open the telephone case
Replace the PSB 2165 with the PSB 2163
Switch the DIP-switch according to table 3
Close the telephone case
Normally, the PSB 2163 is not only used in handset mode but also as a speakerphone,
therefore the speakerphone test mode is of interest, which indicates the internal status
of the speakerphone with the help of two LEDs. As table 1 shows, with the PSB 2163
this internal status information is outputted at the PZ-pins (pin 27, 28). Inside the
SIPB 5132-SP telephone, the piezo-ringer is connected to these pins. With regard to the
schematic of the telephone, the following steps have to be performed for a proper
connection of the test LEDs:
• Adjust the volume potentiometer for the piezo ringer to the lowest volume or
disconnect the piezo ringer (otherwise you hear a "click" each time the LEDs are
turned on or off)
• Connect pin 27 (PZ2) with the gate of (e.g.) T4
• Connect pin 28 (PZ1) with the gate of (e.g.) T5 (see schematic)
Now the pins SA, SB (pin 5 and 4), which are normally connected to the gates of T4 and
T5, must always be programmed as inputs (default) to avoid a short circuit at the PZ pins.
Of course, pin 4 and 5 can also be disconnected. Note, that pin SD is used for hook
switch detection and therefore T2/D1 should not be used for the test LEDs.
Table 3
DIP-switch Inside the SIPB 5132-SP Telephone, Equipped with the PSB 2163
SW1
SW2
Mode
ON
ON
IOM-2 TE, A-Chip
ON
OFF
IOM-2 TE, B-Chip
OFF
ON
IOM-2 NON-TE, A-Chip
OFF
OFF
IOM-2 NON-TE, B-Chip
Semiconductor Group
214
PSB 2163
3
Using the SIPB 5132-SP Telephone with the PSB 2163
in IOM-2 NON-TE Mode
The 4.096 MHz-IOM-2 mode offers the possibility to connect the PSB 2163 directly to a
LineCard-Module or for measuring purposes. Since this IOM-2 mode offers eight
channels, the pins SB, SC, SD serve as slot-select inputs, as mentioned in table 1. A
useful solution is to connect all slot-select pins to VCC (+ 5 V), thus making the
ARCOFI-SP use channel 7. This is also appropriate for the ARCOS-SP PLUS software.
In addition to the steps explained in chapter 2, the best way is to solder pull-up resistors
(4k7) at the pins 2, 3, 4 (SB, SC, SD). Note, that when using the IOM-2 NON-TE mode
the state of the hook switch can not be transmitted in the C/I-channel. The wire between
pin 2 (SD) and the hook switch (SW3) must be disconnected.
Because of the high clock frequency (4 MHz) the flat-band cable between the
SIPB 5132-SP telephone and the rest of the test equipment should be as short as
possible. Tests have shown that the SIPB 5132-SP with its 1m cable can interface with
the SIPB 5130 Audio-Module or the STUT 2000 PERCOFI-Board without any
problems. However, in case of troubleshooting, the length is critical and must be taken
into account.
Semiconductor Group
215
PSB 2163
Changes for the PSB 2163
Semiconductor Group
216
Evaluation Board ARCOFI®-SP V1.0 SIPB 5133-SP
- VAKATSEITE -
SIPB 5133-SP
Evaluation Board ARCOFI®-SP V1.0 SIPB 5133-SP
Table of Contents
Page
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
2
Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
3
Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
4
Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Connector Pin-Outs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Service Access Connector SAC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Audio Interface Connector AIC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
External Power Supply Connector EXC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Handset Connector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Wiring Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
List of Replaceable Parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Floor Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
222
222
222
223
224
224
225
226
226
5.1
5.2
5.3
5.3.1
5.3.2
5.4
5.5
Operational Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Setting Switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Setting Jumper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Configuring the Evaluation Board ARCOFI®-SP . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IOM-2 TE-Mode Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IOM-2 NON-TE-Mode Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AC/DC-Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Connectable Adapter Boards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
227
227
227
227
228
228
228
229
6
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
7
Menu Software Track Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Track File HS_S02_I.TE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Track File LC_1.IOM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Track File HS_LC.TRK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Track File LC_PCM4.TRK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1
4.1.1
4.1.2
4.1.3
4.1.4
4.2
4.3
4.4
5
7.1
7.2
7.3
7.4
Semiconductor Group
219
230
230
232
237
239
SIPB 5133-SP
1
Introduction
The development of analog front-ends for voice transmission in ISDN-terminals poses enormous
efforts to the system houses. The engineers in fulfilling the very strict requirements of the CCITT are
to solve these problems partly by experience and partly by following a trial and error procedure.
The Evaluation Board ARCOFI-SP SIPB 5133-SP helps in developing analog front-ends. It offers
the possibility to experiment on pure analog transmission parts and to investigate the transmission
functions very comfortably in using a PCM4-measuring device by Wandel & Goltermann.
The integration of the Evaluation Board ARCOFI-SP into the SIPB-system with its peculiar software
renders developing and adapting a voice transmission part to simplify connecting a module: No
protocols need be taken notice of and no specific adapter circuitries for testing the ARCOFI-SP are
to be developed any longer.
2
•
•
•
•
•
•
•
•
•
•
3
Features
Comfortable measuring aid for evaluating various transmission functions:
Transfer Functions
Gain Tracking
Signal / Noise Ratio
Group Delay
Enables testing the proper adaptation via programmed FX- and FR-filters of the ARCOFI-SP of
electro-acoustic transducers
Programmable via IOM-2 without need for noticing any protocols
Suitable for IOM-2 TE mode and IOM-2 NON TE mode
All analog inputs and outputs and the digital signals SA, SB, SC, SD of the ARCOFI-SP
connected to the Audio Interface Connector (AIC)
Power supply (+ 5 V) selectable from the PC or from an external network, programmable via one
jumper J1.
Use
The use of the Evaluation Board ARCOFI-SP is identical to that of a standard ISDN-Telephone
Board. Due to the flexible concept of the user board system it can be employed either as an analog
front-end or as a measurement front-end in a terminal equipment (TE) configuration or in a line-card
architecture using the IOM-2 NON TE interface.
The Evaluation Board ARCOFI-SP is always connected to the Audio Interface Module via the IOMinterface. Figure 1 shows, how and in what combination it is connected via a flat cable to the
mainboard.
Semiconductor Group
220
SIPB 5133-SP
Figure 1
Examples of Two Terminal Configurations in the User Board Concept
Semiconductor Group
221
SIPB 5133-SP
4
Circuitry
The circuitry of the Evaluation Board ARCOFI-SP can be divided into a digital and an analog part.
The digital part consists of:
• ARCOFI-SP IC1 (PSB 2163 or PSB 2165)
The analog part is realized in form of an Audio Interface Connector (AIC) where all listed signals are
accessible:
• Analog inputs:
• Analog outputs:
MIN1, MIP1
MIN2, MIP2
MI3
HOP, HON
LSP, LSN
• Reference voltage: VREF
• Digital signals:
SA, SB, SC, SD
• Power supply:
+ 5 V, GND
In addition means for DC blocking is included. By means of the jumper the power supply for the
analog front-end can be chosen: + 5 V from the PC or from a separate, external power supply.
The push button switch SW3 resets the whole circuitry by resetting the ARCOFI-SP. The different
interface modes and the hardware address can be selected by switches S1 and S2 of SW1 for the
ARCOFI-SP.
4.1
Connector Pin-Outs
4.1.1 Service Access Connector SAC
Figure 2
Pin-Outs of the Service Access Connector SAC
Semiconductor Group
222
SIPB 5133-SP
Pin
Function
1
Power supply + 5 V
2
GND
3
n.c.
4
DD
5
n.c.
6
DCL (1.536 MHz / 4.096 MHz)
7
FSC (8 kHz)
8
DU
9
Reset
4.1.2 Audio Interface Connector AIC
Figure 3
Pin-Outs of the Audio Interface Connector AIC
Pin/Row
A
B
C
13
14
15
16
17
18
19
20
VREF
VREF
VREF
SD
SA
+ 5 VA
x
+ 5 VD
GNDA
x
x
SB
+ 5 VA
x
+ 5 VD
GNDA
GNDD
x
SC
LSP
LSN
x
x
GNDD
x = not connected
Semiconductor Group
223
SIPB 5133-SP
4.1.3 External Power Supply Connector EXC
Figure 4
Pin-Outs of the External Power Supply Connector ST1
Pin
Function
1
+5V
2
GND
3
GND
4
X (reserved)
4.1.4 Handset Connector
Figure 5
Handset Connector
Pin
Function
1
MIP
2
HOP
3
HON
4
MIN
Semiconductor Group
224
SIPB 5133-SP
4.2
Wiring Diagram
Figure 6
Wiring Diagram of the Evaluation Board ARCOFI®-SP
Semiconductor Group
225
SIPB 5133-SP
4.3
List of Replaceable Parts
Component
Outline
Type / Value
IC1
R 1, R 2
R3
R4
RP1
P-DIP-28 F
RR1
RR1
RR1
DNET 6
PSB 2165 (replacable with PSB 2163)
100 kΩ
2.7 kΩ
4.7 kΩ
2.7 kΩ
C1, C3, C5, C7, C9
C2, C4, C6, C8, C10
C11
C12
C13
C14
XTANT
CRM 2 + 5
X7R
MKT
XTANT
CRM 2 + 5
1 µF/16 V
100 nF
3.3 nF/16 V
1 µF
10 µF
100 nF
SW1
SW3
SAC
ST1
P-DIP-4
SWITCH
SWITCH
4.4
CAN9
Power Connector
Floor Plan
Figure 7
Floor Plan of the Evaluation Board ARCOFI®-SP
Semiconductor Group
226
SIPB 5133-SP
5
Operational Information
All information about installing the Evaluation Board ARCOFI-SP, programming the ARCOFI-SP
and the other modules of the set-up in various configurations, are described in this chapter.
For programming the set-up using the menu software track files see chapter 7. These
configurations can also be programmed by the ARCOS-SP Plus software. For more details on
ARCOS-SP Plus refer to “ARCOS-SP Plus User Manual V 1.0”.
5.1
Setting Switches
There are three switches at the Evaluation Board ARCOFI-SP. The push button switch SW3 is to
reset the whole circuitry by resetting the ARCOFI-SP. Switches S1 and S2 of SW1 select the
interface mode (IOM-2 TE or IOM-2 NON TE) and the hardware address (AD0 or AD1) of the
ARCOFI-SP respectively. The particular settings are depicted in the table below.
S1
S2
Function
ON
ON
IOM-2 TE mode (A-chip)
ON
OFF
IOM-2 TE mode (B-chip)
OFF
ON
IOM-2 NON TE mode (A-chip)
OFF
OFF
IOM-2 NON TE mode (B-chip)
5.2
Setting Jumper
One jumper J1 is used to select the power supply: If the power is taken from the PC, jumper J1 is
set to position 1-2 (towards SW3). If an external power source is connected via the external power
supply connector ST1, jumper J1 is set to position 2-3 (towards ST1).
5.3
Configuring the Evaluation Board ARCOFI®-SP
The Evaluation Board ARCOFI-SP can be used in two different interface modes:
• IOM-2 TE; terminal timing (1.536 MHz DCL))
• IOM-2 NON TE; non terminal timing (4.096 MHz DCL)
Attention:
When using the non terminal timing mode (IOM-2 NON TE) the timeslot must be determined by
hardwiring the pins SD, SC, SB (pins 2,3,4). By default, these pins are not connected on the
SIPB 5133-SP Evaluation Board. Refer to the User’s Manual, chapter 2.3.2 “IOM-2 Frame
Structure and Timing Modes” for more detailed information.
Before starting the initialization procedure using the menu software, please recheck the
configuration and the jumpers on the modules to succeed in programming. For IOM-2 NON TE
mode, one of the eight timeslot has to be set. Note that the Evaluation Board ARCOFI-SP has
always to be connected to that service access connector of the mainboard that leads to the audio
interface module (see also figure 1).
Semiconductor Group
227
SIPB 5133-SP
5.3.1 IOM-2 TE-Mode Configuration
This configuration consists of:
• a PC-Mainboard
• an Audio Interface Module
SIPB 5000
SIPB 5130
DIP-switches S01, S02 set to ON; S03, S04 set to OFF
• a S-Access Module
SIPB 5100
DIP-switches S01, S02, S03 set to OFF; S04 set to ON
The whole initialization procedure of this configuration is listed in the track file named
"HS_S02_I.TE" in the appendix.
5.3.2 IOM-2 NON-TE-Mode Configuration
This mode needs:
•
•
•
•
a PC-Mainboard
an Audio Interface Module
a LineCard Module
a S-Access Module
SIPB 5000
SIPB 5130
SIPB 5121
SIPB 5100
DlP-switch S02 is set to ON, others OFF (open);
To set this configuration two track files have to be used. The file “LC_1.IOM” provides a general
setup for the EPIC on the LineCard-Module and must be run first. Afterwards either “HS_LC.TRK”
can be used to program the handset mode or “LC_PCM4.TRK” to prepare measurements with the
PCM4 measurement device from Wandel&Goltermann.
Note, that due to the high transmission frequency on the IOM-2 NON TE bus, the length of the flat
band cable between the Audio Interface Module and the Evaluation Board is critical and should not
exceed 0.5 meters.
5.4
AC/DC-Characteristics
Important for a noiseless voice transmission and ideal test results are:
• A big ground area underneath the ARCOFI-SP, where digital ground GNDD and analog ground
GNDA are connected.
• Blocking capacitors at every IC and at the entry of the evaluation board.
• The flat cable, which connects the ARCOFI-SP Evaluation Board with the audio interface
module, has to be of a big cross section and of a good isolation.
• The analog signal lines on the adapter boards should be well screened.
Attention:
CMOS-lCs are very sensible on electrostatic discharge. Never pull out the SAC-plug at the
ARCOFI-SP Evaluation Board before switching off the power supply.
Semiconductor Group
228
SIPB 5133-SP
5.5
Connectable Adapter Boards
To the Evaluation Board ARCOFI-SP various adapter boards can be connected via the Audio
Interface Connector AIC.
• In connection with the Loudhearing Adapter Board SIPB 5133 LH the evaluation board meets all
basic functions for voice transmission. On this board there are four pins for connecting a German
type handset, and a special connector for joining a US American type handset. In addition the
signals LSN and LSP are accessible for connecting a loudspeaker.
6
Glossary
AIC
ARCOFI
DTMF
EPIC
EXC
FHM
FR
FX
GNDA
GNDD
GND – 5 VA
GND + 5 VA
HAC
HOP, HON
HS
IOM
ISDN
LSP, LSN
MIP, MIN
PCM4
PIC
SAC
SIPB
TE
XINP, XINN
VA
VD
Semiconductor Group
Audio Interface Connector
Audio Ringing COdec Filter
Dual Tone Multi Frequency
Extended PCM Interface Controller PEB 2055
EXternal power supply Connector
Free Hand Microphone
Filter in Receive direction
Filter in Transmit direction
Ground Analog
Ground Digital
Ground for the – 5 V Analog Voltage
Ground for the + 5 V Analog Voltage
Hands-free Add-on Circuit
Handset Output Positive / Negative
Handset
ISDN-Oriented Modular
Integrated Services Digital Network
Loudspeaker Output Positive / Negative
Microphone Input Positive / Negative
Measurement Device for PCM channels from Wandel&Goltermann
PC-Interface Connector
Service Access Connector
Siemens ISDN PC-User Board System
Terminal Equipment
Auxiliary Input Positive / Negative
Voltage Analog
Voltage Digital
229
SIPB 5133-SP
7
Menu Software Track Files
7.1
Track File HS_S02_I.TE
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
W
W
R
R
R
C
C
C
C
C
W
R
R
W
R
C
C
C
C
C
R
A
C
C
C
C
**********************************************************************
**********************************************************************
*
* Trackfile HS_S02_I.TE
*
* Programming the ARCOFI-SP PSB 2163 in HANDSET Mode
* and additionally activating the S0 interface
*
* !! COSI /2 MUST BE RUN BEFORE !!
*
**********************************************************************
**********************************************************************
---------------------------------------------------------------------ISAC-S in IOM2 mode
---------------------------------------------------------------------/S02TE/ISAC_S/SERIAL/ADF2 80
/S02TE/ISAC_S/SERIAL/SPCR 80
/S02TE/ISAC_S/HDLC/STAR 4A
/S02TE/ISAC_S/HDLC/STAR 4A
/S02TE/ISAC_S/HDLC/STAR 4A
---------------------------------------------------------------------Activation of the S0 - interface
---------------------------------------------------------------------/S02TE/ISAC_S/SERIAL/CIX0
/S02TE/ISAC_S/SERIAL/CIR0
/S02TE/ISAC_S/SERIAL/CIR0
/S02TE/ISAC_S/SERIAL/SPCR
/S02TE/ISAC_S/SERIAL/SPCR
60
30
30
05
05
---------------------------------------------------------------------IOM2 - Identification of ARCOFI-SP
---------------------------------------------------------------------/S02TE/ISAC_S/BUS/MON1 A000 01
A084
---------------------------------------------------------------------Programming the ARCOFI-SP in HANDSET mode
----------------------------------------------------------------------
Semiconductor Group
230
SIPB 5133-SP
D
B
B
B
B
B
X
C
C
C
C
D
B
B
X
C
C
C
C
D
B
X
W
C
C
C
C
C
C
R
A
C
C
C
C
C
C
C
C
A11F
0060
4100
0020
F112
/S02TE/ISAC_S/BUS/MON1
---------------------------------------------------------------------COP_6: GZ=-15dB
---------------------------------------------------------------------A126
9932
/S02TE/ISAC_S/BUS/MON1
---------------------------------------------------------------------Powering up the ARCOFI-SP
---------------------------------------------------------------------A131
/S02TE/ISAC_S/BUS/MON1
/S02TE/ISAC_S/SERIAL/CIX1 20
---------------------------------------------------------------------Read Out CR: 00, 60, 41, 00
00, 20, F1, 16
---------------------------------------------------------------------/S02TE/ISAC_S/BUS/MON1 A19F 05
A100 6041 0000 20F1 16FF
**********************************************************************
**********************************************************************
*
*
End of Track File
*
**********************************************************************
**********************************************************************
Semiconductor Group
231
SIPB 5133-SP
7.2
Track File LC_1.IOM
C **********************************************************************
C
LC_1.IOM
C **********************************************************************
C
C application: initialization of the line card module
C
for a SICOFI2/ARCOFI-SP measurement
C setup:
line card module SIPB 5121
C
C iom2 channel assignement:
C * ch0: analog subscriber (sicofi2)
C * ch1: analog subscriber (sicofi2)
C * ch2: analog subscriber (sicofi2)
C * ch3: analog subscriber (sicofi2)
C * ch4: analog subscriber (sicofi2)
C * ch5: analog subscriber (sicofi2)
C * ch6: analog subscriber (sicofi2)
C * ch7: analog subscriber (sicofi2)
C
C interface characteristics:
C pcm interface: 2 hws with 32 ts each
C cfi interface: 4 hws with 32 ts each
C
(4 iom2 interfaces)
C
C configuration of the lc module:
C config register bits:
C id,cks/tc2/tc1/tc0/dch/dma/cts/res
C * clock mode 6 (xtal 4096khz)
C * reset of on board devices
W /LINECA/CONFIG/CONFIG/CONFIG 61
W /LINECA/CONFIG/CONFIG/CONFIG 60
C
C configuration of the pcm interface:
C * pcm mode 0
W /LINECA/EPIC/PCMCFI/PMOD 20
W /LINECA/EPIC/PCMCFI/PCSR 11
W /LINECA/EPIC/PCMCFI/POFD F1
W /LINECA/EPIC/PCMCFI/POFU 19
C
C configuration of the cfi interface:
C * cfi mode 0, clock source: pcl/pfs
C * pfs evaluated with falling edge
C * prescaler = 1
W /LINECA/EPIC/PCMCFI/CMD1 20
C * fsc output: fc mode 6
C * dcl output: double rate
Semiconductor Group
232
SIPB 5133-SP
C
W
C
W
C
W
C
W
C
C
C
W
C
C
C
W
C
C
W
W
C
C
C
W
C
C
C
C
C
W
W
W
C
W
W
C
W
W
C
W
W
C
C
C
C
C
W
* xmit rising, rec falling edge
/LINECA/EPIC/PCMCFI/CMD2 D0
* cfi bit number is 256
/LINECA/EPIC/PCMCFI/CBNR FF
* pfs marks cfi ts31,bit1
/LINECA/EPIC/PCMCFI/CTAR 02
* no shift between xmit and rec
/LINECA/EPIC/PCMCFI/CBSR 00
* subchannel position:64kbps=bits7.0
32kbps=bits7.4
16kbps=bits7.6
/LINECA/EPIC/PCMCFI/CSCR 00
initialization of cm ctrl field:
* cm reset mode
/LINECA/EPIC/MARSCR/OMDR 00
* ff is copied to all positions of
* the cm ctrl field
/LINECA/EPIC/MARSCR/MADR FF
/LINECA/EPIC/MARSCR/MACR 70
cfi configuration for iom2:
* cm init mode
/LINECA/EPIC/MARSCR/OMDR 80
cfi timeslots 2 and 3 of port 0
are programmed as monitor and
signaling channels (analog iom)
* ts 2 downstream:
/LINECA/EPIC/MARSCR/MADR FF
/LINECA/EPIC/MARSCR/MAAR 08
/LINECA/EPIC/MARSCR/MACR 7A
* ts 3 downstream:
/LINECA/EPIC/MARSCR/MAAR 09
/LINECA/EPIC/MARSCR/MACR 7B
* ts 2 upstream:
/LINECA/EPIC/MARSCR/MAAR 88
/LINECA/EPIC/MARSCR/MACR 7A
* ts 3 upstream:
/LINECA/EPIC/MARSCR/MAAR 89
/LINECA/EPIC/MARSCR/MACR 7A
cfi timeslots 6 and 7 of port 0
are programmed as monitor and
signaling channels (analog iom)
* ts 6 downstream:
/LINECA/EPIC/MARSCR/MADR FF
Semiconductor Group
233
SIPB 5133-SP
W
W
C
W
W
C
W
W
C
W
W
C
C
C
C
C
W
W
W
C
W
W
C
W
W
C
W
W
C
C
C
C
W
W
W
C
W
W
C
W
W
C
W
W
C
C
C
/LINECA/EPIC/MARSCR/MAAR
/LINECA/EPIC/MARSCR/MACR
* ts 7 downstream:
/LINECA/EPIC/MARSCR/MAAR
/LINECA/EPIC/MARSCR/MACR
* ts 6 upstream:
/LINECA/EPIC/MARSCR/MAAR
/LINECA/EPIC/MARSCR/MACR
* ts 7 upstream:
/LINECA/EPIC/MARSCR/MAAR
/LINECA/EPIC/MARSCR/MACR
18
7A
19
7B
98
7A
99
7A
cfi timeslots 10 and 11 of port 0
are programmed as monitor and
signaling channels (analog iom)
* ts 10 downstream:
/LINECA/EPIC/MARSCR/MADR FF
/LINECA/EPIC/MARSCR/MAAR 28
/LINECA/EPIC/MARSCR/MACR 7A
* ts 11 downstream:
/LINECA/EPIC/MARSCR/MAAR 29
/LINECA/EPIC/MARSCR/MACR 7B
* ts 10 upstream:
/LINECA/EPIC/MARSCR/MAAR A8
/LINECA/EPIC/MARSCR/MACR 7A
* ts 11 upstream:
/LINECA/EPIC/MARSCR/MAAR A9
/LINECA/EPIC/MARSCR/MACR 7A
cfi timeslots 14 and 15 of port 0
are programmed as monitor and
signaling channels (analog iom)
* ts 14 downstream:
/LINECA/EPIC/MARSCR/MADR FF
/LINECA/EPIC/MARSCR/MAAR 38
/LINECA/EPIC/MARSCR/MACR 7A
* ts 15 downstream:
/LINECA/EPIC/MARSCR/MAAR 39
/LINECA/EPIC/MARSCR/MACR 7B
* ts 14 upstream:
/LINECA/EPIC/MARSCR/MAAR B8
/LINECA/EPIC/MARSCR/MACR 7A
* ts 15 upstream:
/LINECA/EPIC/MARSCR/MAAR B9
/LINECA/EPIC/MARSCR/MACR 7A
cfi timeslots 18 and 19 of port 0
are programmed as monitor and
signaling channels (analog iom)
Semiconductor Group
234
SIPB 5133-SP
C
W
W
W
C
W
W
C
W
W
C
W
W
C
C
C
C
W
W
W
C
W
W
C
W
W
C
W
W
C
C
C
C
W
W
W
C
W
W
C
W
W
C
W
W
C
C
* ts 18 downstream:
/LINECA/EPIC/MARSCR/MADR FF
/LINECA/EPIC/MARSCR/MAAR 48
/LINECA/EPIC/MARSCR/MACR 7A
* ts 19 downstream:
/LINECA/EPIC/MARSCR/MAAR 49
/LINECA/EPIC/MARSCR/MACR 7B
* ts 18 upstream:
/LINECA/EPIC/MARSCR/MAAR C8
/LINECA/EPIC/MARSCR/MACR 7A
* ts 19 upstream:
/LINECA/EPIC/MARSCR/MAAR C9
/LINECA/EPIC/MARSCR/MACR 7A
cfi timeslots 22 and 23 of port 0
are programmed as monitor and
signaling channels (analog iom)
* ts 22 downstream:
/LINECA/EPIC/MARSCR/MADR FF
/LINECA/EPIC/MARSCR/MAAR 58
/LINECA/EPIC/MARSCR/MACR 7A
* ts 23 downstream:
/LINECA/EPIC/MARSCR/MAAR 59
/LINECA/EPIC/MARSCR/MACR 7B
* ts 22 upstream:
/LINECA/EPIC/MARSCR/MAAR D8
/LINECA/EPIC/MARSCR/MACR 7A
* ts 23 upstream:
/LINECA/EPIC/MARSCR/MAAR D9
/LINECA/EPIC/MARSCR/MACR 7A
cfi timeslots 26 and 27 of port 0
are programmed as monitor and
signaling channels (analog iom)
* ts 26 downstream:
/LINECA/EPIC/MARSCR/MADR FF
/LINECA/EPIC/MARSCR/MAAR 68
/LINECA/EPIC/MARSCR/MACR 7A
* ts 27 downstream:
/LINECA/EPIC/MARSCR/MAAR 69
/LINECA/EPIC/MARSCR/MACR 7B
* ts 26 upstream:
/LINECA/EPIC/MARSCR/MAAR E8
/LINECA/EPIC/MARSCR/MACR 7A
* ts 27 upstream:
/LINECA/EPIC/MARSCR/MAAR E9
/LINECA/EPIC/MARSCR/MACR 7A
cfi timeslots 30 and 31 of port 0
are programmed as monitor and
Semiconductor Group
235
SIPB 5133-SP
C
C
W
W
W
C
W
W
C
W
W
C
W
W
C
R
C
C
C
W
R
R
C
C
C
C
W
W
C
C
C
C
C
W
C
C
W
C
C
C
C
C
C
C
C
C
C
signaling channels (analog iom)
* ts 30 downstream:
/LINECA/EPIC/MARSCR/MADR FF
/LINECA/EPIC/MARSCR/MAAR 78
/LINECA/EPIC/MARSCR/MACR 7A
* ts 31 downstream:
/LINECA/EPIC/MARSCR/MAAR 79
/LINECA/EPIC/MARSCR/MACR 7B
* ts 30 upstream:
/LINECA/EPIC/MARSCR/MAAR F8
/LINECA/EPIC/MARSCR/MACR 7A
* ts 31 upstream:
/LINECA/EPIC/MARSCR/MAAR F9
/LINECA/EPIC/MARSCR/MACR 7A
* pcm status is
/LINECA/EPIC/MARSCR/STAR 05
* not synchronized (pss=0)
setting epic to normal mode
/LINECA/EPIC/MARSCR/OMDR C0
/LINECA/EPIC/MARSCR/ISTA 08
/LINECA/EPIC/MARSCR/STAR 25
pcm status: synchronized (pss=1)
initialization of the pcm tristate
field, all ch. to high impedance
/LINECA/EPIC/MARSCR/MADR 00
/LINECA/EPIC/MARSCR/MACR 68
activation epic:
* normal mode, pcm and cfi active
* cfi output drivers push-pull
* mf ch. handshake protocol enabled
/LINECA/EPIC/MARSCR/OMDR E6
reset cififo:
/LINECA/EPIC/MARSCR/CMDR 10
*********************************************************************
the line card is now ready for use with the SICOFI2 or the ARCOFI-SP
Now you can run your trackfile with the filter coefficients or
program the ARCOFI-SP
*********************************************************************
end of trackfile
*********************************************************************
Semiconductor Group
236
SIPB 5133-SP
7.3
Track File HS_LC.TRK
C **********************************************************************
C **********************************************************************
C
C Track File HS_LC.TRK
C
C Programming the ARCOFI-SP PSB 2163
C via IOM-2 NON-TE in Handset Mode
C
C !! Trackfile LC_1.IOM has to be run before !!
C
C Configuration:
C ------------C LineCard
SIPB 5121
C Audio Module SIPB 5130
C
DIP-Switch 1 ON
C
2 ON
C
3 OFF
C
4 ON
C
C **********************************************************************
C **********************************************************************
C
C------------------------------------------------------------------C Connecting the LineCard to the Audio Interface Module
C------------------------------------------------------------------C
W /LINECA/CONFIG/CONFIG/CONFIG E0
C
C------------------------------------------------------------------C Selecting the timeslot
C (depends on the pins SB,SC,SD)
C Here: slot 7 (all pins to VCC)
C
C MFSAR = 04 => slot 0
C MFSAR = 0C => slot 1
C MFSAR = 14 => slot 2
C MFSAR = 1C => slot 3
C MFSAR = 24 => slot 4
C MFSAR = 2C => slot 5
C MFSAR = 34 => slot 6
C MFSAR = 3C => slot 7
C------------------------------------------------------------------C
W /LINECA/EPIC/MCHSTR/MFSAR 3C
C
Semiconductor Group
237
SIPB 5133-SP
C------------------------------------------------------------------C ARCOFI-SP Identification
C------------------------------------------------------------------W /LINECA/EPIC/MCHSTR/CMDR 01
W /LINECA/EPIC/MCHSTR/MFFIFO A0
W /LINECA/EPIC/MCHSTR/MFFIFO 00
W /LINECA/EPIC/MCHSTR/CMDR 08
R /LINECA/EPIC/MCHSTR/ISTA 20
R /LINECA/EPIC/MCHSTR/STAR 26
R /LINECA/EPIC/MCHSTR/MFFIFO A0
R /LINECA/EPIC/MCHSTR/STAR 26
R /LINECA/EPIC/MCHSTR/MFFIFO 84
W /LINECA/EPIC/MCHSTR/CMDR 01
C
C---------------------------------------------------------------------C COP_6: GZ=-15dB
C------------------------------------------------------------------W /LINECA/EPIC/MCHSTR/MFFIFO A1
W /LINECA/EPIC/MCHSTR/MFFIFO 26
W /LINECA/EPIC/MCHSTR/MFFIFO 99
W /LINECA/EPIC/MCHSTR/MFFIFO 32
W /LINECA/EPIC/MCHSTR/MFFIFO 00
W /LINECA/EPIC/MCHSTR/MFFIFO 00
W /LINECA/EPIC/MCHSTR/CMDR 04
W /LINECA/EPIC/MCHSTR/CMDR 01
C
C------------------------------------------------------------------C SOP_F: Handset Mode
C------------------------------------------------------------------W /LINECA/EPIC/MCHSTR/MFFIFO A1
W /LINECA/EPIC/MCHSTR/MFFIFO 1F
W /LINECA/EPIC/MCHSTR/MFFIFO 00
W /LINECA/EPIC/MCHSTR/MFFIFO 60
W /LINECA/EPIC/MCHSTR/MFFIFO 41
W /LINECA/EPIC/MCHSTR/MFFIFO 00
W /LINECA/EPIC/MCHSTR/MFFIFO 00
W /LINECA/EPIC/MCHSTR/MFFIFO 20
W /LINECA/EPIC/MCHSTR/MFFIFO F1
W /LINECA/EPIC/MCHSTR/MFFIFO 16
W /LINECA/EPIC/MCHSTR/CMDR 04
W /LINECA/EPIC/MCHSTR/CMDR 01
C
C**********************************************************************
C*
C* End of Track File
C*
C **********************************************************************
Semiconductor Group
238
SIPB 5133-SP
7.4
Track File LC_PCM4.TRK
C **********************************************************************
C
C Track File LC_PCM4.TRK
C
C Preparing the ARCOFI-SP PSB 2163 for PCM4 measurements
C
C Trackfile LC_1.IOM has to
C be run before !
C
C Configuration:
C ------------C LineCard
SIPB 5121
C Audio Module SIPB 5130
C
DIP-Switch 1 ON
C
2 ON
C
3 OFF
C
4 ON
C PCM4 Adapter SIPB 5311
C
Jumper open
C
C **********************************************************************
C
C---------------------------------------------------------------------C Connecting the LineCard to the Audio Interface Module
C---------------------------------------------------------------------C
W /LINECA/CONFIG/CONFIG/CONFIG E0
C
C---------------------------------------------------------------------C Selecting the timeslot (depends on the pins SB,SC,SD)
C Here: slot 7 (all pins to VCC)
C---------------------------------------------------------------------C
W /LINECA/EPIC/MCHSTR/MFSAR 3C
C
C---------------------------------------------------------------------C B-channel switching for the PCM4 Adaptor
C B1 -> TS1
C B2 -> TS2
C---------------------------------------------------------------------C
W /LINECA/EPIC/MARSCR/MADR 0F
W /LINECA/EPIC/MARSCR/MAAR 81
W /LINECA/EPIC/MARSCR/MACR 60
C
Semiconductor Group
239
SIPB 5133-SP
W /LINECA/EPIC/MARSCR/MADR 0F
W /LINECA/EPIC/MARSCR/MAAR 88
W /LINECA/EPIC/MARSCR/MACR 60
C
W /LINECA/EPIC/MARSCR/MADR 81
W /LINECA/EPIC/MARSCR/MAAR F0
W /LINECA/EPIC/MARSCR/MACR 71
C
W /LINECA/EPIC/MARSCR/MADR 01
W /LINECA/EPIC/MARSCR/MAAR 70
W /LINECA/EPIC/MARSCR/MACR 71
C
W /LINECA/EPIC/MARSCR/MADR 88
W /LINECA/EPIC/MARSCR/MAAR F1
W /LINECA/EPIC/MARSCR/MACR 71
C
W /LINECA/EPIC/MARSCR/MADR 08
W /LINECA/EPIC/MARSCR/MAAR 71
W /LINECA/EPIC/MARSCR/MACR 71
C
C---------------------------------------------------------------------C ARCOFI-SP Identification
C---------------------------------------------------------------------C
W /LINECA/EPIC/MCHSTR/CMDR 01
W /LINECA/EPIC/MCHSTR/MFFIFO A0
W /LINECA/EPIC/MCHSTR/MFFIFO 00
W /LINECA/EPIC/MCHSTR/CMDR 08
R /LINECA/EPIC/MCHSTR/ISTA 20
R /LINECA/EPIC/MCHSTR/STAR 26
R /LINECA/EPIC/MCHSTR/MFFIFO A0
R /LINECA/EPIC/MCHSTR/STAR 26
R /LINECA/EPIC/MCHSTR/MFFIFO 84
W /LINECA/EPIC/MCHSTR/CMDR 01
C
C---------------------------------------------------------------------C Insert now the desired programming sequence for the PSB 2163
C---------------------------------------------------------------------C
C**********************************************************************
C**********************************************************************
C*
C* End of Track File
C*
C**********************************************************************
C**********************************************************************
Semiconductor Group
240
ARCOFI®-SP Coefficients Software ARCOS-SP
and ARCOS-SP PLUS SIPO 2163 V1.0
SIPO 2163
Table of Contents
Page
1
1.1
1.2
1.3
1.4
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The ARCOFI®-SP PSB 2163 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The ARCOS-SP PLUS Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
System Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Installation and Activation of ARCOS-SP PLUS . . . . . . . . . . . . . . . . . . . .
244
244
244
245
246
2
2.1
2.2
2.2.1
2.2.2
2.2.3
2.2.4
2.2.5
2.2.6
2.3
2.4
Using ARCOS-SP PLUS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Menu Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Using the Menu Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pull-Down Menu "File" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pull-Down Menu "ARCOFI" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pull-Down Menu "Show" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pull-Down Menu "Board" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pull-Down Menu "Options" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The User Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
248
248
249
249
249
251
251
251
252
252
255
3
3.1
3.2
3.3
3.4
3.4.1
3.4.2
Generating the Correction Filter Coefficients . . . . . . . . . . . . . . . . . . . .
Filter Implementation and Theory of Calculation . . . . . . . . . . . . . . . . . . . .
Calculation of Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Reading the FX- and FR-Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Examples for the Usage of the FX/FR Filters . . . . . . . . . . . . . . . . . . . . . . .
Adapting the FX Filters to a Target Frequency Response . . . . . . . . . . . . .
Using the "Execute" Feature to Calculate Coefficients . . . . . . . . . . . . . . .
259
259
260
265
265
265
267
4
4.1
4.2
4.3
Hardware Setup for Use with ARCOS-SP PLUS . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Using the SIPB 5000 System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Other Hardware Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
269
269
269
273
5
5.1
5.2
Error Messages and Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Hardware Related Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Other Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
Semiconductor Group
243
SIPO 2163
1
Introduction
1.1
The ARCOFI®-SP PSB 2163
The PSB 2163 provides the subscriber with an optimized Audio, Ringing, Codec, Filter
processor solution for a digital telephone. The ARCOFI-SP fulfils all the necessary
requirements for a full-featured digital telephone including handsfree operation. The
ARCOFI-SP performs all coding, decoding and filtering according to CCITT and ETSI
standards. The outstanding advantage of the PSB 2163 is the high performance
speakerphone implementation with the "stronger-wins algorithm" that allows almost a full
duplex conversation.
Due to the completely digital concept of the circuit no external components are required.
The various filters and processing steps of the voice channel including the
speakerphone, the signalling, and the tone generation can be adapted to meet the
different operating conditions by means of software. Also adapting different acoustical
transducers to the ARCOFI-SP is accomplished simply by programming registers and
coefficients.
Therefore evaluating the features and the high performance of the PSB 2163 is an easy
thing if some software is available to program the chip. An important task of the
ARCOS-SP PLUS software is to offer an quick and easy way for evaluation.
1.2
The ARCOS-SP PLUS Software
The ARCOS-SP PLUS program has been designed to generate all the required
coefficients and to program the ARCOFI-SP in a real environment. Besides the software
offers an easy way to get familiar with the internal structure of the ARCOFI-SP because
all functional blocks are displayed in a graphic form with many possibilities of interaction
(switches, parameters, registers that can be clicked upon).
Note: The ARCOS-SP PLUS software is also available as a demonstration software.
The demo-version is called ARCOS-SP (instead of ARCOS-SP PLUS). This
user’s manual describes ARCOS-SP PLUS but is also used as a manual for
ARCOS-SP. However, certain commands are not possible with the demo-version
and cause an error message ("ARCOS-SP does not support this command...").
The demonstration software ARCOS-SP does not support any kind of hardware access.
That means, it is neither possible to load or save files nor to access the ARCOFI-SP with
the help of the SIPB userboards or any other hardware. Because this restriction is easily
understood, there will be no further hints throughout this manual that describe, what
things can not be done with the demo-version. The calculation of all the coefficients is
possible with ARCOS-SP since this requires no hardware access.
Semiconductor Group
244
SIPO 2163
Features of the ARCOS-SP PLUS Software
ARCOS-SP PLUS supports the calculation and programming of
•
•
•
•
•
•
Coefficients for the ARCOFI-SP digital speakerphone
Coefficients for the three ARCOFI-SP tone generation registers
Coefficients for the ARCOFI-SP DTMF tone generator registers
Coefficients for both ARCOFI-SP programmable gain registers GX and GR
Coefficients for the GZ side tone gain register
Coefficients for the FX and FR correction filter registers;
adaptive software calculates coefficients to fit a target frequency response
ARCOS-SP PLUS supports also the configuration registers and offers an user-friendly
dialogue mode allowing full programming of the ARCOFI-SP configuration registers and
the coefficient RAM (CRAM).
Other features simplify working with the ARCOFI-SP:
•
•
•
•
Access to different kinds of hardware (see chapter 4 for details)
NOTE and EXECUTE (a kind of keyboard macro)
READ ARCOFI and WRITE ARCOFI
Support of transmission measurements using the Wandel&Goltermann PCM4
measuring instrument
It is highly recommended to have an PSB 2163 User’s Manual at hand when working
with the ARCOS-SP PLUS software. It is recommended to use the manual as a
reference when working with ARCOS-SP or ARCOS-SP PLUS software. On the other
hand, not all coefficients are documented in detail in the User’s Manual of the PSB 2163
and the ARCOS-SP PLUS or the ARCOS-SP software allow easy determination of all
coefficients.
1.3
System Requirements
An IBM or compatible Personal Computer (AT or better) is required. The computer
should have at least 400 kByte of free conventional DOS memory. DOS version 3.2 or
newer is required. The usage of a mouse as an input device is recommended and offers
access to all features of the user area. A mouse driver must be installed before calling
the ARCOS-SP PLUS software.
For hardware access the SIEMENS ISDN PC Board (SIPB) system is required. The
SIPB family consists of a mainboard (SIPB 5000) and a wide variety of modules to
realize different applications. The following equipment is needed:
•
•
•
•
•
•
•
A Mainboard SIPB 5000
Any Layer-1 module SIPB 511x
Any Layer-2 module SIPB 512x
An Audio Interface Module SIPB 5130, EPROM version 2.0 or newer
As well as one of the following modules containing the ARCOFI-SP:
An ARCOFI-SP Telephone SIPB 5132-SP1)
An ARCOFI-SP Evaluation Board SIPB 5133-SP
Semiconductor Group
245
SIPO 2163
The SIPB 5133-SP is an ARCOFI-SP Evaluation Board used for measurement
purposes. The SIPB 5132-SP is an ARCOFI-SP Telephone used for demonstration and
teaching purposes.
The ARCOS-SP PLUS software not only supports the SIPB 5000 system, but also some
evaluation boards that have serial interfaces. These are:
• SIPB 8051 ISDN Telephone and Terminal Adapter Development Board,
• SISI 2197 SmartLink Board,
• STUT 2000 PERCOFI-Board.
The setups for the different kinds of hardware are described in chapter 4.
1.4
Installation and Activation of ARCOS-SP PLUS
Installation
The following simple procedure is recommended for the installation of ARCOS-SP PLUS
on a hard-disc:
Create ARCOS-SP directory: md c:\ARCOS-SP
Change to ARCOS-SP directory: cd ARCOS-SP
Copy ARCOS-SP files: copy a:*.* c:\ARCOS-SP\*.*
At least the following files should have been copied:
•
•
•
•
ARC63.EXE, the main program (ARC63D.EXE for the ARCOS-SP software)
ARC63.TAB, data file (ARC63D.TAB for the ARCOS-SP software)
RS232.INI, initialization file for the serial interface
MSHERC.COM (hercules driver; only necessary if the system is equipped with a
Hercules monochrome graphics card)
The program MSHERC.COM loads a Hercules driver resident in the RAM. When using
a Hercules graphics card, this program must be called before starting
ARCOS-SP PLUS.
The file ARC63.INI (ARC63D.INI) is generated by ARCOS-SP itself, it is not delivered
with the program.
1
These telephones are equipped with an PSB 2165 which can be replaced with an PSB 2163
(refer to the App. Note "Using the SIPB 5132-SP Telephone with the PSB 2163")
Semiconductor Group
246
SIPO 2163
Starting ARCOS-SP PLUS
Enter the following DOS command to start ARCOS-SP:
ARC63 [File[.xxx]]
[File[.xxx]] is an optional initialization file. The initialization file contains information
about the last hardware settings and the options chosen (see chapter 4). As default, the
file ARC63.INI is used as initialization file.
Please note, that the ARCOS-SP PLUS software must be started with a special option
when using external hardware which is controlled via the serial interface (command line
switch /V; this is described in chapter 4.3).
After having started ARCOS-SP PLUS the following is carried out:
•
•
•
•
Loading the initialization file,
Loading the coefficient file,
Showing the ARCOS-SP PLUS main menu and the SIEMENS label (once a day),
After a mouse click or carriage return the SIEMENS label will disappear.
The next step to be done is either to disable any hardware access or to initialize the
hardware.
Semiconductor Group
247
SIPO 2163
2
Using ARCOS-SP PLUS
2.1
Introduction
When working with ARCOS-SP PLUS the screen is divided into four main areas (please
compare with figure 1):
•
•
•
•
Row 1 is reserved for the menu line
Rows 2 to 23 are for the User Area
Row 24 contains the command line
Row 25 is the status line.
All items that can be activated via the menu line are described in chapter 2.2. The user
area is controlled with a mouse and allows to program the whole ARCOFI-SP
(chapter 2.3). Everything that can be done with the mouse in the user area can also be
done via the command line using the keyboard. The syntax for the command line can be
found in chapter 2.4.
Figure 1
ARCOS-SP PLUS Screen
Semiconductor Group
248
SIPO 2163
2.2
The Menu Line
2.2.1
Using the Menu Line
The menu items can always be activated by pressing either the F10-key or the ALT-key
in combination with the highlightened letter of the menu item (e.g. "ALT+O" for the
options menu). Of course, clicking with the left mouse button is also possible. The
subsequent paragraphs describe each menu item.
2.2.2
Pull-Down Menu "File"
"Load"
The name of the file to be loaded has to be entered in the input field or can be selected
in the file list. The command to load a file will be aborted if an incorrect, non-existent file
name is entered. Both kinds of files: *.ARC and *.ARB are loadable (refer to the next
paragraph). This menu item is used to restore a previously saved set of coefficients.
The load command can also be activated by pressing "Ctrl+L".
"Save"
The complete programming of the ARCOFI-SP as well as the state of the ARCOS-SP
PLUS software can be saved in a file.
When saving, a file name and format is asked for. There are two different formats
supported by ARCOS-SP PLUS. They are an ARCOS-SP PLUS specific binary file
format and a text format. If no extension is given, the entered name will automatically
receive the following one:
• Sequence of commands: *.ARC (text format)
• State of ARCOS-SP PLUS: *.ARB (binary format)
Afterwards the filename is checked for validity. If the file name is not correct, an error
message appears and the save command is aborted.
The binary format is used to store the device status. Since the format is device specific
it is not readable by the user. Being in binary format, the information exchange between
the hard-disc and the ARCOS-SP PLUS software is speeded up.
The text format takes more time to program a sequence since this format has to be
translated to the binary format via the ARCOS command interpreter before it is sent to
the ARCOFI-SP. The user, however, can read this format, which facilitates programming
the ARCOFI-SP. For documentation purposes the contents of an *.ARC file can be used
in standard word-processing software. It can even be altered with an ASCII-Editor and
then again be loaded with the LOAD command.
The save command can also be activated by pressing "Ctrl+S".
Semiconductor Group
249
SIPO 2163
"Execute"
A file name is asked for and then checked for validity. Any file which was previously
recorded with the "Note" function, may be read and executed. If the file name is invalid
or non-existent, an error message is shown and the execute command is aborted.
"Note"
It is possible to record all activities between the ARCOFI-SP and ARCOS-SP PLUS. The
feature "Note" can be turned on or off to record all actions in a pre-named file.
To implement "Note" the user has to activate the pull-down menu. When "Note" is active,
the menu command will have a check mark to the left of it. Once active, one has to name
the file and give an appropriate extension to it. The file extension depends on what one
wants to record, either a sequence of commands or just a line input.
• Extension *.ARC : ARCOFI-SP sequence of commands: only ARCOFI-SP
programming
• Extension *.ARS : ARCOS-SP PLUS commands; a record of every key stroke or
mouse activity
The *.ARS file consists of a list of all commands that the user has entered via the
ARCOS-SP PLUS dialogue mode to program the ARCOFI-SP. It is possible to read the
*.ARC file and review what was programmed. The *.ARS commands, however, are
written in an ARCOS-SP PLUS specific program language to program the ARCOFI-SP.
This feature allows to run a test pattern several times without looking at the file, thereby
saving time.
"DOS Shell"
"DOS Shell" is provided for calling the operating system. It is possible to toggle back and
forth between ARCOS-SP PLUS and the DOS command level. The following actions are
carried out:
• The screen in ARCOS-SP PLUS is completely stored as it is,
• The DOS operating system is called,
• When returning to ARCOS-SP PLUS, the screen reappears just the way it was left
before.
"Exit"
To leave the ARCOS-SP PLUS program, use the "Exit" command or just type a "Q" in the
command line.
Semiconductor Group
250
SIPO 2163
2.2.3
Pull-Down Menu "ARCOFI"
"READ ARCOFI"
The command READ ARCOFI reads the complete device. If the coefficients of the
ARCOS-SP PLUS software are not identical to the coefficients in the ARCOFI-SP, the
current window as well as the status line are corrected.
This command can also be activated by pressing "Ctrl+R".
"WRITE ARCOFI"
With the WRITE ARCOFI function the ARCOS-SP PLUS software coefficients which are
displayed on the screen are written to the ARCOFI-SP. In general, this is always done
automatically by the software, but the WRITE ARCOFI command allows to write all
coefficients at once a second time.
This command can also be activated by pressing "Ctrl+W".
2.2.4
Pull-Down Menu "Show"
This feature enables the user to switch between various windows in the user area to
show the device parameters that are read from the chip and to manipulate them. The
following windows can be displayed (refer to chapter 2.3):
• CRAM (Coefficient RAM)
• Register
• ARCOFI
– AFE (Analog Front End)
– ADI (ARCOFI Digital Interface)
– ASP (ARCOFI Signal Processor)
– Tone-Generator
– DTMF-Generator
– Speakerphone
– SD-Transmit (Speech Detector)
– SD-Receive (Speech Detector)
– SC-Acoustic (Speech Comparator for the acoustic echo)
– SC-Line (Speech Comparator for the line echo)
2.2.5
Pull-Down Menu "Board"
This menu serves for changing the configuration of the Mainboard, setting the signalling
channel and choosing the ARCOFI-SP device (interface mode and chip address).
Semiconductor Group
251
SIPO 2163
"A-Chip/B-Chip"
These two menu items can only be selected when the IOM-2 mode is activated. In this
case switching between two devices via software is possible. If the serial interface mode
was selected (SPI Serial Programming Interface; SCI/SDI) this feature is not available.
"Signalling"
This menu listing allows to set the bits in the C/I channel 1. Please note, that this menu
item is only available in the IOM-2 TE mode of the ARCOFI-SP. The setting of the bit
CAM in the GCR-register influences the bit positions in the signalling window (refer to the
ARCOFI-SP User’s Manual, chapter 3.4.3).
2.2.6
Pull-Down Menu "Options"
A variety of options exists with the ARCOS-SP PLUS software. It is possible to set the
colors, the sensitivity of the mouse, the beeper (ON/OFF), the tone frequency of the error
beep, the double borders for the menu (ON/OFF) and the shadow effect (for the
windows).
2.3
The User Area
Working with the User Area
The user area is the portion of the screen between row 2 and 23 (see figure 2.2). In this
area different background items can be displayed and manipulated with the mouse. The
screen contents of the user area can be chosen either with the menu item "Show" from
the main menu or simply by clicking at different boxes inside the user area with the left
mouse button.
Being in the user area of ARCOS-SP PLUS, only a click with the left button of the mouse
will activate the different fields.
A field can have three different states:
• ON: the border and contents of the field are highlighted
• OFF: the border and contents of the field are dimly displayed
• ERROR: the border and contents of the field are displayed in red
The left and right mouse buttons have different meanings. The left mouse button is
used to program a field or used to zoom into another window. A double click with the left
button turns a field ON/OFF and at the same time programs the corresponding bit in the
configuration register. The right mouse button is used to read a field and to show the
coefficients in hexadecimal code, i.e. as they have to be written into the ARCOFI-SP.
Semiconductor Group
252
SIPO 2163
Initializing the Hardware
Before it is possible to work with the user area either the external hardware must be
initialized (click at the field "Initialize Hardware", compare with figure 2.2) or it must be
disabled (field "Disabled"). With a disabled hardware the ARCOS-SP PLUS software
can be used to evaluate ARCOFI-SP coefficients without influencing any external
hardware containing the ARCOFI-SP. If the hardware is successfully initialized, any
programming action will also concern the ARCOFI-SP.
The hardware is initialized as soon as the initialization field in the ARCOFI-SP window is
clicked upon or when the ARCOFI-SP is accessed. Access takes place either with a
"Read" command or with a "Load" command. In this case the hardware will be
programmed completely with the ARCOS-SP PLUS conditions displayed on the screen.
In chapter 4 a description of the different setups depending on the external hardware
can be found.
Changing Registers and Coefficients in the User Area
Functions or switches can be activated simply by clicking upon the desired, double
framed fields. This is equivalent to the bit combinations in the registers as they can be
found in the register window. For example the sidetone gain stage GZ can be activated
by setting the GZ bit in the register window or simply with a double click at the GZ box in
the user area. Another example is the DTMF bit. This bit can be activated with a single
click on the DTMF switch in the user area.
A coefficient like the sidetone gain coefficient required for GZ can be programmed with
a single click at the GZ box in the user area. As a result, a pop-up window appears which
offers all possible values. Should the desired value not be shown, one can scroll through
the list by clicking upon the small up and down arrows located at the top of the frame
surrounding the coefficient values. It is also possible to type in the desired value. The
program automatically chooses the closest value available. If the value of the coefficient
is unknown, then two question marks will be shown. The entered coefficient must carry
the same unit as the main header of the window.
These explanations apply to all bits and coefficients. In general, the user area is selfexplanatory and can also be used to become familiar with the architecture of the
ARCOFI-SP. However, the generation of coefficients for the FX- and FR-filters is a bit
more difficult to understand and will be explained separately in chapter 3.
Semiconductor Group
253
SIPO 2163
The Register Window and the CRAM Window
For experienced users, the two windows "Register" and "CRAM" offer the possibility to
change almost every bit or coefficient of the ARCOFI-SP with the help of only two
different screen masks. Everything that can be done in the other windows of the user
area can be done with the register and CRAM window instead. Using the right mouse
button in the register and CRAM window shows the complete programming sequence
that is required for the actual setting of the ARCOFI-SP.
Clicking upon register bits makes them toggle or a pop-up window appears. CRAM
coefficients can also be altered by clicking upon them. A double click at one of the
coefficients causes ARCOS-SP PLUS to switch to the corresponding window in the user
area. For example a double click at the parameter ATT in the CRAM window activates
the speakerphone window.
Semiconductor Group
254
SIPO 2163
2.4
The Command Line
Introduction
It is highly recommended to use a mouse to work with the ARCOS-SP PLUS software,
but it is also possible to enter all inputs via the keyboard. All commands have to be
entered in the command line (see figure 12.2).
This chapter gives a complete description of all inputs that can be done in the command
line. The command line is not case sensitive. The F3 key can be used to recall the last
command.
ARCOS-SP PLUS Commands
Show
ASP | Reg[ister] | ARCOFI | AFE | ADI |
CRAM | Tone[generator]| DTMF[generator] |
Spe[akerphone] | SDT[ransmit] | SDR[eceive]
W[rite]
ARCOFI
R[ead]
ARCOFI
Board
{ A[-][Chip] | B[-][Chip]} | {B1 | B2} | Sig[nalling]
[<Hex><Hex>] | En[abled] | Dis[abled] | Init[ialise]}
Load
<Filename>
Save
<Filename>
Note
<Filename>
Exec[ute] <Filename>
Delay
<Time>
DOS
Exit
Q[uit]
Commands for Setting Coefficients
<Command> <Coefficient>
[<Coefficients Set>]
DTMF
<Float><Float>[<Float><Float>] F3, G3, [FD, GD3]
F[T]
<Float> <Float>[<Float>]
F1, F2, [F3]
FS
<Float> <Float>[<Float>]
F1S, F2S, [F3S]
G
<Float> <Float>[<Float>]
G1, G2, [G3]
GD
<Float> <Float>[<Float>]
GD1, GD2, [GD3]
T
<Float> <Float>[<Float>]
T1, T2, [T3]
FX
<OptimMode> <Files> <OptimSpeed>
FR
<OptimMode> <Files> <OptimSpeed>
Semiconductor Group
255
SIPO 2163
Coefficients
GX
| GR
| GZ
TOn
| TOff
F1[T]
| F2[T]
| F3[T]
F1S
| F2S
| F3S
G1
| G2
| G3
GD1
| GD2
| GD3
T1
| T2
| T3
A1
| A2
| K
| GE
GAE
| GLE
| ATT
| ETAE
TW
| DS
| SW
GDSAE
| PDSAE
| GDNAE
| PDNAE
GDSLE
| PDSLE
| GDNLE
| PDNLE
LIM
| OFFX
| OFFR
LP2LX
| LP2LR
| LP1X
| LP1R
PDSX
| PDNX
| PDSR
| PDNR
LP2SX
| LP2NX
| LP2SR
| LP2NR
LGAX
| LGAR
COMX
| AGX
| TMHX
| TMLX
| NOISX
COMR
| AGR
| TMHR
| TMLR
| NOISR | AAR
| FD
Extended Commands
Po[wer] Down
Po[wer] Up
Re[set]
Semiconductor Group
256
| ETLE
SIPO 2163
Configuration Setting Commands
S[et]
<CRBit> [0 | 1 | I | II | ALw | uLw | in | out | SQ | TR]
C[lr]
<CRBit>
S[et]
VDM <Hex>
C[lr]
VDM
S[et]
MIC|AMI
C[lr]
MIC|AMI
S[et]
AIMX|AIN[-MUX]
C[lr]
AIMX|AIN[-MUX]
S[et]
HOC|AHO
C[lr]
HOC|AHO
S[et]
LSC|ALS
C[lr]
LSC|ALS
S[et]
DLTF
C[lr]
DLTF
S[et]
Tone[generator]
C[lr]
Tone[generator]
S[et]
Spe[akerphone]
C[lr]
Spe[akerphone]
S[et]
AGCX
C[lr]
AGCX
S[et]
AGCR
C[lr]
AGCR
S[et]
Comp|Exp|VDM
C[lr]
Comp|Exp|VDM
S[et]
{ GCR | DFICR | PFCR | TGCR | TGSR | ATCR | ARCR | TFCR |
[down | by [-pass]| 0 | 6 | 12 | 18 | 24 |
30 | 36 | 42]
[MIP1[|MIN1]|MIP2[|MIN2]|MI3]
[down | by [-pass] | 2.5 | -3.5 | -9.5 |
-15.5 | -21.5]
[down | by [-pass] | 11.5 | 8.5 | 5.5 |
2.5 | -0.5 | -3.5 | -6.5 | -9.5 | -12.5 |
-12.5 | -15.5 | -18.5 | -21.5 ]
[ NOT | IDR | DLP | DLS | DLN ]
SDICR | XCR } <Hex><Hex>
C[lr]
GCR | DFICR | PFCR | TGCR | TGSR | ATCR | ARCR | TFCR |
SDICR | XCR
Semiconductor Group
257
SIPO 2163
Low-Level Setting Commands
W[rite] [0|1]
<WCommands>
R[ead]
<RWCommands>
Print Commands
P[rint]
DTMF | CRAM | AFE
| ADI
|Tone[generator]|
Spe[akerphone]| AG[CX] | AG[CR] | Reg[ister] |
GX | GR | GZ | Exp[&VDM] | Comp
P[rint]FX Freq[uency]|Res[ult]|<OptimMode> <Files> <OptimSpeed>
Command Parameters
<CRBit>
SP | AGCX | AGCR | EVX | SLOT | PU | CAM | ESIG |
LAW | SD | SC | SB | SA | VDM3 | VDM2 | VDM1 | VDM0 |
GX | GR | GZ | FX | FR |DHPR | DHPX | TG | DT | ETF |
CG | BT | BM | SM | SQTR | PM | TRL | TRR | DTMF |
TRX | MIC3 | MIC2 | MIC1 | MIC0 | EVREF | AIMX1 |
AIMX0 | HOC2 | HOC1 HOC0 | CME | LSC3 | LSC2 | LSC1 |
LSC0 | DHS | EPZST | ALTF2 | ALTF1 | ALTF0 | DLTF2 |
DLTF1 | DLTF0 | EPP | DCE | MCLKR2 | MCLKR1 | MCLKR0 |
PGCR | PGCX | RAAR | OBS | DHOP | DHON | DLSP | DLSN
<RWCommands>
SOP_0
SOP_4
SOP_8
COP_0
COP_4
COP_8
COP_C
|
|
|
|
|
|
|
<WCommands>
XOP_0
XOP_E
| XOP_1
| XOP_F
<Files>
<Filename> [<Filename> [<Filename>
[<Filename>]]]
<Filename>
Filename with extension
<OptimMode>
F[lat] | T[arget] | L[imited]
<OptimSpeed>
/F[ast]| /M[iddle]| /B[est]
<Float>
Standard Real Number
<Time>
<Float>
<Hex>
0 ... F
Semiconductor Group
258
SOP_1
SOP_5
SOP_A
COP_1
COP_5
COP_9
COP_D
|
|
|
|
|
|
|
SOP_2
SOP_6
SOP_D
COP_2
COP_6
COP_A
COP_E
|
|
|
|
|
|
|
| XOP_D
|
SOP_3
SOP_7
SOP_F
COP_3
COP_7
COP_B
COP_F
|
|
|
|
|
|
SIPO 2163
3
Generating the Correction Filter Coefficients
3.1
Filter Implementation and Theory of Calculation
Two high performance frequency correction filters FX and FR are implemented in the
ARCOFI-SP, allowing an optimum adaption to different types of transducers or
compensating the frequency response of the telephone plastics itself. Specifications of
different countries can be fulfilled by means of a simple software change.
One filter consists of two equalizers with variable gain, factor of quality, and center
frequency followed by a high-/low-pass filter. The filters can only attenuate the signal,
they behave like passive filters. The FX or FR filter is adjusted with the help of twelve
coefficients which are calculated by the ARCOS-SP PLUS software.
It is obvious, that there are different ways to configure the two equalizers and the high-/
low-pass filter to achieve one and the same frequency response. Therefore always more
than one coefficient set is capable of fulfilling the desired response.
The basic idea for calculating the coefficients is, that two frequency responses are given.
One represents the desired over-all frequency response (target function), the other one
is the frequency response, that the hardware actually offers (input function). While the
first one is usually a flat frequency response, the latter one can be measured or
estimated. ARCOS-SP PLUS then calculates the filter coefficients (filter function) in a
way, that the filter function plus the input function results in the target function. If for
example the hardware transmits high frequencies only attenuated, it exhibits a kind of
low-pass function (input function). If a flat frequency response is desired (target function)
ARCOS-SP PLUS would calculate the coefficients in a way, that the FX or FR filter
shows a high-pass behavior in order to compensate the input function.
Furthermore, it is not only possible to give a desired target function but also an upper and
lower limit for the target function can be given.
The algorithm used for generating the filter coefficients is based on a stochastic method,
similar to the law of cooling: i.e. at the beginning of the process, the parameters (here the
filter coefficients) may be within a large range. This range progressively is reduced such
that the parameters converge. Up to and including the 8th calculation step the quality of
the approximation is defined by the mean square of the difference between the target
function and the addition of the filter functions of FX or FR and the input function. Thus
the discrepancy between the two curves is calculated for each sampling frequency, then
it is squared, added and finally divided by the number of the sampling frequency points.
From the 9th calculation step the quality of the approximation is defined by the greatest
absolute difference. The algorithm shows the value of the absolute minimum and the
quality of the coefficient set which is presently considered as the best solution.
In order to calculate the discrepancy properly, each amplitude curve is normalized.
Before the algorithm starts, each amplitude curve is shifted such that it has an amplitude
Semiconductor Group
259
SIPO 2163
value of 0 dB at one sampling frequency point. The reference frequency is the sampling
frequency which is equal to or closest to 1000 Hz.
Note: Because the algorithm is based on a random process, different runs of the
program, starting from the same input function, can give different results. It is
recommended to make several runs and to take the best result (see also
page 262).
3.2
Calculation of Coefficients
The coefficients for the FX transmit and FR receive path correction filters can be
calculated either with the ARCOS-SP PLUS or with the ARCOS-SP program. Both
programs generate filter coefficients and graphically display the filter transfer functions
(amplitude transfer function in dB, group delay response in µs). The graphic output is
only possible if the system is equipped with an EGA, VGA, CGA, or a Hercules graphics
card. The transfer function can also be stored in tabular form in two separate files, one
for the amplitude response (*.AMP) and one for the group delay response (*.GDY).
After clicking at the FX or FR filter box in the ASP window (ARCOFI-SP signal processor)
an dialogue box appears that offers the choice between three basic optimization modes.
For each mode the optimization speed must be chosen as either "fast", "middle" or
"best". Once the optimization mode, the selected speed of optimization and the required
file names have been entered, the calculation starts.
Optimization Speed
The optimization process might require long computing time in some difficult cases, in
particular a "best" fit with over 100 frequency points might take hours to be completed
when using a slow PC without co-processor. It has been found that between 20 to 30
sampling points usually give a satisfactory result while keeping the calculation time
reasonably short.
Optimization Mode "flat target function"
For a flat frequency response, the algorithm approximates the inverted function of the
input function. The input function must be declared in an input file. The file has to be
entered with the filename and the extension. An example for an input file shows table 1.
For the FR filter a second input function can be entered (e.g. one for the handset
earpiece, the second for the loudspeaker). The calculation points of this second input
function are interpolated and subsequently averaged. Hereby ARCOS-SP PLUS will
eliminate automatically the frequency areas which are not common in both input
functions (ranges at lower and upper frequencies).
When the coefficients have been calculated, the inverse curve of the input function is
displayed if a graphics screen is available. On the same graph the filter functions of FX
Semiconductor Group
260
SIPO 2163
or FR are displayed. Both curves are shifted such that they have the same amplitude
value at the reference frequency.
After having quit (using pressing carriage return or the key "Q") the discrepancy between
the two curves is displayed. The graphics mode can be left by quitting a second time.
Figure 2
Screen after Calculating Filter Coefficients
The optimization result is displayed in a separate window after successful completion
(see figure 2 for an example). Here the values GR- or GX-Adjustment, r.m.s. deviation
and linear deviation are displayed as well as the relevant input files. Because the FX/FR
filters can be regarded as passive filters, the additional loss caused by the filter must be
compensated. ARCOS-SP PLUS displays different values that represent this additional
loss and are also a measure for the quality of the optimization:
• "lin. deviation" is the minimum filter loss
• "Receive Adjustment" ("Transmit Adjustment") is the filter loss weighted according to
CCITT and ETSI NET33
• "GR-Adjustment" ("GX-Adjustment") is the weighted filter loss over the complete
frequency range of 4 kHz
• "r.m.s. deviation" is the mean square of the difference between the filter function and
the desired response
Semiconductor Group
261
SIPO 2163
Figure 3 illustrates the meaning of these values.
Figure 3
Illustration of Filter Loss and Deviation
Especially the "linear deviation" is a measure for the quality of the coefficient set. The
lower this value, the better the approximation is. Several runs of the program can result
in different values for the linear deviation because of the random starting parameters the
algorithm uses. In general, it should be possible to reach a value of less than 3dB for the
"lin. deviation".
Optimization Mode "target function"
The algorithm tries to compensate the input function in such a way that the addition of
this input function and of the filter functions of FX or FR is as close as possible to the
target function. The target function must be given in a file.
Before starting the algorithm, the sampling frequency points of the target function are
adjusted to the sampling frequency points of the input function. ARCOS-SP PLUS
checks if each sampling frequency point of the input function has a corresponding
sampling frequency point in the target file. If a sampling frequency point is not found in
the file, then the amplitude value of this sampling frequency point is interpolated by
making a linear approximation from the two points closest to the desired sampling
frequency point. Thus the algorithm can work with two curves not having the same
sampling frequency points.
Semiconductor Group
262
SIPO 2163
If a sampling frequency point is higher than the highest point or lower than the lowest
point of the target function then the algorithm considers that the frequency response can
have any value at this point. These calculation sampling points are not recognized and
therefore do not influence the calculation.
After the calculation of the coefficients the input functions and filter functions of FX or FR
are added and the result is graphically displayed. In the same graph the target function
is also shown. Both curves are shifted such that they have the same amplitude value at
the reference frequency.
After quitting the discrepancy, i.e. the difference between the two curves, is graphically
shown. When quitting again one leaves the graphics mode. Finally the results as
described for the "flat frequency response" are displayed.
Optimization Mode "limited by an upper and lower curve"
The algorithm tries to compensate the input function such that the addition of this input
function and of the filter functions of FX or FR is as close as possible to the middle of the
upper and the lower limiting curves (which is then the target function). Before starting the
algorithm, the target function as well as the weighting factors are calculated.
First, the sampling frequency points of the upper limit and lower limit are adjusted to the
ones of the input function. Then the algorithm checks whether each sampling frequency
point of the input function has a corresponding sampling frequency point in the upper and
lower function. If a sampling frequency point is not found in the upper and lower curves,
then the amplitude value for this sampling frequency point is interpolated by making a
linear approximation from the two points closest to the desired sampling frequency point.
Finally for each frequency point the mean of the upper and the lower limit is calculated.
These mean values describe the target function.
If a sampling frequency point is higher than the highest point or lower than the lowest
point of the upper and lower limiting curve, then the algorithm considers that the
frequency response can have any value at this point. These calculation samples would
not be recognized.
The ratio of the longest and shortest distance between the upper and the lower limiting
curves (for one frequency) defines the size of a weighting factor. The shorter the
distance between the limiting curves, the better the correction, the larger the weighting
factor. The weighting factor is taken into consideration when calculating the greatest
absolute difference as well as the mean of the square of the difference.
After the calculation of the coefficients, the input function and the filter functions of FX or
FR are added and the result is graphically displayed. In the same graph the upper and
lower limits are also shown. The curves are shifted in such a way, that the input function
and the target function have the same amplitude value at the reference frequency.
Finally the mean square of the difference and the other values described above, are
displayed.
Semiconductor Group
263
SIPO 2163
File Format
A maximum of up to 100 frequency/amplitude points can be used by the approximation
algorithm. The frequency points must be in progressing order between 1 Hz and
3999 Hz. The amplitude values should be limited to a maximum of |20| dB although this
is not required for the fitting algorithm.
The software expects one pair of values (frequency and level) in one line, both values
separated by one or more blanks and/or tab-stops. The first value must be the frequency,
the second one the level. Units are Hz and dB; they are not part of the file. Numbers can
be integer or real (e.g. 3 or - 3.1 are allowed). Table 1 shows an example for an input file.
Table 1
Example for an Input File
Frequency (Hz) Level (dB)
300
−1
700
−1
800
− 0.5
1000
0
1050
0.1
1200
0.5
1250
0.75
1300
1
1350
1.25
1400
1.5
1450
1.75
1500
2
1700
3.75
1750
4.25
1900
5.5
2250
6.9
2550
7.45
3000
8
3250
6.75
3350
6.25
Semiconductor Group
264
SIPO 2163
3.3
Reading the FX- and FR-Filters
A click with the right mouse button (or the input Print FX or Print FR) at the
command line) invokes the read window for the filters. This window is a blend between
an input and an output window. There are three fields which can be clicked upon and one
entry field to enter a file name. If one clicks upon the field "Frequency Response", then
the frequency and phase response of the filter are shown in a graphical format. If one
chooses the field "Optimization", the three values of the last calculation and the file name
are shown (optimization result window). With a click on the "Check Optimization" field,
the same window is displayed that appears after programming the filters (see figure 2),
but the coefficients are not recalculated, instead the result of the optimization is taken. If
it is desired that the frequency response is saved in a separate file, the file name has to
be entered in this window.
3.4
Examples for the Usage of the FX/FR Filters
3.4.1
Adapting the FX Filters to a Target Frequency Response
The use of the FX filter to match a required mask template will be highlighted in the
following example.
According to the European Telecommunications Standards Institute (ETSI) standard,
the sending sensitivity and frequency response from the mouth reference point (MRP) to
the digital interface shall be within a mask template given in table 2.
Table 2
Mask Template According to ETSI
Frequency (Hz)
Upper Limit (dB)
Lower Limit (dB)
100
− 12
200
0
300
0
− 12
1000
0
−6
2000
4
−6
3000
4
−6
3400
4
−9
3999
0
The sampling points (in units of Hz and dB) of the upper and the lower limits are written
into a text file with the help of an editor. The first column contains the frequency values
in Hz, the second column contains amplitude values in dB (refer also to table 1). The
units themselves are not inputs, they are generated by ARCOS-SP PLUS.
Semiconductor Group
265
SIPO 2163
The names chosen for the files in this example are UPPER.LIM and LOWER.LIM.
Table 3
Listing of Boundary Curves (in Hz and dB)
Upper Curve
Lower Curve
300
− 12
1000
0
1000
−6
2000
4
3000
−6
3400
4
3400
−9
The frequency response of the microphone can be measured. The sampling points are
also input into a text file. For example it is assumed, that the appropriate file is
PRIMO.DAT as shown in table 1. Because frequencies under 300 Hz and above
3400 Hz are blocked by the ARCOFI-SP filters, only the sampling points in the range
300 Hz - 3400 Hz should be entered in the file. In order to obtain a good approximation,
at least 20 points between 300 Hz and 3400 Hz should be selected. However, not more
than 100 points should be entered.
After having entered the measurement points, the program ARCOS-SP PLUS is called.
When in the ASP-Window, a click with the left mouse button upon the FX field opens a
display where the input files, the optimization mode and the optimization speed have to
be selected. For this example the optimization mode "limited by an upper and a lower
curve" must be chosen. The input function is the file called PRIMO.DAT, the upper and
lower curves are given with the files UPPER.DAT and LOWER.DAT. A click upon the "OK"
field starts the calculation.
At the end of the approximation process the frequency response of the microphone
(input function) and that of the FX correction filter (filter function) are added and the sum
is displayed together with the two limiting curves. Notice that the curves are shifted. The
value of the amplitude at the reference frequency (1000 Hz in the example) is 0 dB. The
two limiting curves are shifted such that the mean value of the limiting curves at the
reference frequency is 0 dB.
As already described, the linear deviation of the filter function is a indication of the quality
of the coefficient set. The calculation in this example should be done several times until
a set of coefficients is found, that results in a deviation of less than 2 dB.
Semiconductor Group
266
SIPO 2163
3.4.2
Using the "Execute" Feature to Calculate Coefficients
The process of starting the filter calculation several times and checking the results after
each calculation can be eased by using the "Execute" function of ARCOS-SP PLUS.
With the help of an appropriate *.ARS file, the software calculates as many sets of
coefficients as desired and allows to check the optimization result afterwards.
To understand the contents of the *.ARS file it is necessary to remember the command
line syntax of ARCOS-SP PLUS, as far as it is required for the FX/FR filters:
FX <OptimMode> <Files> <OptimSpeed>
FR <OptimMode> <Files> <OptimSpeed>
<OptimMode>
:
F[lat]
<OptimSpeed>
:
/F[ast] | /M[iddle] | /B[est]
<Files>
: <DATFile> [ <DATFile>
| T[arget]
[ <LIMFile> [
| L[imited]
<LIMFile> ]]]
<DATFile> : Filename containing an input function
<LIMFile> : Filename containing an upper/lower limit
A simple example illustrates the use of the "Execute" function. It is assumed that the FX
filter should compensate the frequency response given with a file called PRIMO.DAT.
The desired response is "flat" and the optimization mode is "best". The following steps
have to be performed:
• With the help of an ASCII editor a file with the extension *.ARS has to be prepared,
that contains two lines of text for each set of coefficients to be calculated; one line tells
the ARCOS-SP PLUS software to do the calculation, the second line is used to save
the result in an *.ARC file; figure 4 shows the contents of such an *.ARS file, it is
called CALC.ARS; the combination of "FX...SAVE" must be repeated as many times
as sets of coefficients are to be calculated.
• In a second step an *.ARS file has to be prepared to check the previously stored
*.ARC files; see figure 5 for an example, here the file is called CHECK.ARS
• Now the ARCOS-SP PLUS software is started and the menu item FILE, EXECUTE is
chosen; the name of the first *.ARS file is given (e.g. CALC.ARS) and the calculation
starts; depending on the optimization speed and the number of coefficient sets, this
can take quite a long time but the calculation is performed automatically and needs no
interaction from the keyboard.
Semiconductor Group
267
SIPO 2163
• When the calculation is completed, the previously saved *.ARC files can be loaded
separately by using the "Load" command, or the second *.ARS file (CHECK.ARS) is
used to load all the files one after the other and to inspect especially the "lin.
deviation"; the best approximation is taken as a final result then.
FX
Flat primo.dat /B
SAVE
result01.ARC
FX
Flat primo.dat /B
SAVE
result02.ARC
FX
Flat primo.dat /B
SAVE
result03.ARC
Figure 4
Contents of the File CALC.ARS (calculation for 3 sets of coefficients shown)
LOAD result01.ARC
Print FX Result
LOAD result02.ARC
Print FX Result
LOAD result03.ARC
Print FX Result
Figure 5
Contents of the File CHECK.ARS
Semiconductor Group
268
SIPO 2163
4
Hardware Setup for Use with ARCOS-SP PLUS
4.1
Introduction
The ARCOFI-SP PSB 2163 is able to work in three different interface modes:
• IOM-2 TE interface (1.536 MHz)
• IOM-2 NON-TE interface (4.096 MHz)
• Serial control / serial data interface (SCI/SDI)
The ARCOS-SP PLUS supports each of these three interface modes in conjunction with
appropriate hardware. This chapter describes the different hardware setups. First the
ARCOFI-SP has to be connected to the particular hardware which must be configured
correctly. Then the ARCOS-SP PLUS software is called and informed about the
hardware configuration. With a click at the "Initialize Hardware" box the hardware is
checked and initialized for the use with ARCOS-SP PLUS. The ARCOFI-SP itself can be
located either on the SIPB 5133-SP evaluation board, inside a SIPB 5132-SP
telephone, or on a dedicated board like the SIPB 8051 telephone board. Also custom
specific boards with an IOM-2 interface are suitable.
In the "Show ARCOFI" window in the user area of ARCOS-SP PLUS, the following
modes can be set if the hardware support is enabled (compare with figure 1):
• Either the ICC/ISAC as a layer 1/2 device, or the EPIC/ELIC
• Either IOM interface mode, or SPI interface mode (SCI/SDI interface)
• Either A-Chip or B-Chip (only possible in IOM-2 interface mode)
These settings must correspond to the hardware that is connected to the PC on which
the ARCOS-SP PLUS software runs.
4.2
Using the SIPB 5000 System
The ARCOS-SP PLUS software requires a specific SIPB 5000 hardware environment.
The firmware EPROM version 1.1 or newer must be installed on the SIPB 5000
mainboard. Three standard configurations are described in the next paragraphs. More
detailed information can be found in the technical description of the SIPB 5000 system.
Please note, that if the ICC-B is used as a layer-2 device, EPROM version 2.2 or newer
is required for the SIPB 5000 mainboard.
IOM-2 TE Interface with an ISAC-X
A common setup uses the SIPB 5100 (ISAC-S Module) or the SIPB 5103 (ISAC-P
Module) as layer 1/2 module. The ARCOFI-SP is connected via a SIPB 5130 (Audio
Module). Figure 6 shows the general setup either with an ISAC-S or with an ISAC-P
Module. The ARCOFI-SP is located inside a SIPB 5132-SP telephone, but instead of the
telephone any hardware which offers an IOM-2 interface can be connected.
Semiconductor Group
269
SIPO 2163
For the use with ARCOS-SP PLUS the fields ICC/ISAC as well as IOM must be
activated.
Figure 6
Setup for IOM-2 TE Mode with the ISAC-X in TE-Mode
The ISAC-X runs in TE-mode and the B channels are switched to the S0 bus or the UP0
bus respectively. Therefore the setup in figure 6 can be connected to an NT simulator
consisting of another SIPB 5000 mainboard with another ISAC-X in LT-S mode and a
voice connection can be established.
Semiconductor Group
270
SIPO 2163
IOM-2 NON-TE Interface with an EPIC/ELIC
The 4.096 MHz IOM-2 NON-TE interface can also be used if a Linecard-Module
SIPB 5121 is connected to the SIPB 5000 mainboard. Optionally, this configuration
offers the possibility to do measurements with the PCM4 measuring instrument from
Wandel&Goltermann which has to be connected with the help of an PCM4 adapter
SIPB 5311. Figure 7 shows the setup with the ARCOFI-SP placed on the
SIPB 5133-SP ARCOFI-SP evaluation board. Due to the relatively high clock
frequencies it is necessary to keep the cable length between the mainboard and the
evaluation board (or any other hardware platform containing the ARCOFI-SP) as short
as possible. This also applies to the cable between mainboard and PCM4 adapter.
Table 4
Timeslot Assignment for Measurements with the PCM4
Timeslot PCM4
IOM-2 Channel
TS1
B1
TS2
B2
Since the IOM-2 NON-TE interface offers eight channels, it is important to know that the
ARCOS-SP PLUS software supports channel 7 (the last one) and therefore the
ARCOFI-SP must be pin-strapped to this channels. This can easily be done by
connecting1) pull-up resistors to the pins used for timeslot select (SB, SC, SD). For
measurements with the PCM4 measurement device the B1 and B2 channel are switched
to timeslot 1 and 2 (table 4).
1
The resistors must be soldered manually; refer also to the App. Note "Using the SIPB 5132-SP
telephone with the PSB 2163"
Semiconductor Group
271
SIPO 2163
Figure 7
Setup for IOM-2 NON-TE Mode with the Linecard Module
Semiconductor Group
272
SIPO 2163
4.3
Other Hardware Tools
The ARCOS-SP PLUS software is able to communicate with different evaluation boards
via the serial interface port of the PC. The following boards are supported:
• PERCOFI-Board STUT 2000; the ARCOFI-SP is programmed with the help of an
EPIC in IOM-2 TE mode
• Smart Link Kit SISI1097; the SCI/SDI interface of the ARCOFI-SP is used
• ISDN Telephone Board SIPB 8051; the ARCOFI-SP is programmed by an ISAC-X in
IOM-2 TE mode
In order to enable the software to communicate with the boards, a special firmware
version is required for the evaluation boards. This firmware must be loaded to the target
hardware before the ARCOS-SP PLUS software is started. Table 5 shows the names of
the HEX-files with the special firmware versions and the names of the batch files used to
download the HEX-files. The download itself is performed by the utility SIMPLV24.EXE
which is called from the batch files.
Table 5
Hardware to be Connected to the Serial Interface
Hardware Platform
Required
Firmware
Batch File for
Download
Required
RS232 Cable
PERCOFI-Board
STUT 2000
PERCOFI.HEX
LPERCOFI.BAT
Null-modem
Smart Link Kit SISI1097
SMART.HEX
LSMART.BAT
Serial 1:1
ISDN Telephone Board
SIPB 8051
SIPB51.HEX
LSIPB.BAT
Null-modem
Table 6
Configuration to be made in the User Area of ARCOS-SP PLUS
Hardware Platform
Hardware Switches in the User Area
PERCOFI-Board STUT 2000
EPIC/ELIC
Smart Link Kit SISI1097
ISDN Telephone Board SIPB 8051
Semiconductor Group
IOM
SPI
ICC/ISAC
273
IOM
SIPO 2163
Therefore the procedure for starting ARCOS-SP PLUS for the use with one of the
hardware platforms given in table 5 is the following:
• Connect the hardware with an appropriate cable to the PC (the cable is delivered
together with the hardware).
• Make sure that the correct serial port is chosen (default: COM1, see paragraph
"choosing the COM port" for more information).
• Make sure that the hardware is in "loader mode" (see board documentation; switching
between "loader mode" and "program" is performed with the reset button on the
board).
• Download the firmware (batch file according to table 5).
• Make sure, that the board is switched to "program mode"; this is done automatically
after the download, except for the Smart Link Kit where the reset button must be
pressed once (the loader LED on the SmartLink board has to be inactive then).
• Start ARCOS-SP PLUS in the serial interface mode; this is done by adding the switch
/V to the command line input; the syntax is:
ARC [IniFile[.xxx]] /V
When started with this option, the ARCOS-SP PLUS software ignores any SIPB 5000
based hardware and scans the serial port for the presence of one of the boards from
table 5 instead.
• After quitting the hardware message ("board xxx found at the serial port") the usual
window appears in the user area; in this window the hardware switches must be set
according to table 5 before the "initialize hardware" field has to be activated.
Figure 8
Setup for the SIPB 8051 Telephone Board
Semiconductor Group
274
SIPO 2163
The figures 8, 9, and 10 show the setup for the different kinds of hardware. With the
SIPB 8051 Board the B channels are switched to the S interface as well as with the
Smart Link Kit the B channels are transferred over the U interface. Therefore in
conjunction with an NT-simulator, a voice connection can be established.
Figure 9
Setup for the Smart Link Kit
The STUT 2000 PERCOFI Board is mainly intended for measuring purposes and can
directly be connected to the PCM4 measurement device from Wandel&Goltermann, but
nevertheless it can be used to provide an IOM-2 interface for the SIPB 5132-SP
telephones equipped with the PSB 2163. The timeslot assignment for measurements
with the PCM4 measurement device is given in table 7.
Please note, that especially together with the STUT 2000 board, the programming of all
the coefficients over the serial interface takes significantly longer than with the
SIPB 5000 system. When the ARCOS-SP PLUS software accesses the serial port, the
message "RS232 active" appears in the status line.
Semiconductor Group
275
SIPO 2163
Figure 10
Setup for the STUT 2000 PERCOFI Board
Table 7
Timeslot Assignment for Measurements with the STUT 2000 Board and the PCM4
Timeslot PCM4
IOM-2 Channel
TS1
B1
TS2
B2
TS3
IC1
TS4
IC2
Choosing the COM Port
There are two different programs that have to access the serial port. First of all the
ARCOS-SP PLUS software itself, but also the SIMPLV24.EXE utility for downloading
the firmware.
The serial port number for ARCOS-SP PLUS is defined in the RS232.INI file. This file
can be altered with the help of an ASCII editor (default is COM1).
The serial port number for the download of the firmware is given as a command line
argument for SIMPLV24.EXE. If no option for the COM port is given, the utility uses
COM1, otherwise the option /P<n> must be added, where <n> is the number of the
COM port. For this purpose it is convenient to edit the corresponding batch file (see
table 5) and to add e.g. the option /P2 for COM2.
Semiconductor Group
276
SIPO 2163
5
Error Messages and Troubleshooting
This chapter does not offer an complete overview of all error messages since most of
them are self explanatory. Instead, it gives some hints to locate the problems that are
most likely to occur.
5.1
Hardware Related Problems
Message: "Configuration of the Mainboard not supported"
ARCOS-SP PLUS has tried to initialize the hardware but is not able to recognize the
hardware configuration that is currently set with the switches in the user area and/or with
the /V switch in the command line.
Check the configuration of the external hardware (including DIP-switches and jumpers);
does the setting in the user area match with the hardware configuration?; reset the
hardware (reset button), terminate ARCOS-SP PLUS and try it again.
Message: "No ARCOFI found"
The hardware is successfully initialized but it is not possible to establish a connection
with the ARCOFI-SP.
Check the connection between hardware and ARCOFI-SP (e.g. the IOM-2 cable
between the SIPB 5000 mainboard and the SIPB 5133-SP Evaluation Board). Is the
ARCOFI-SP in the correct interface mode (DIP-switches on the SIPB 5133-SP
Evaluation Board)? Is the IOM-2 cable defect or too long? Again a good idea is to reset
the hardware, terminate the software and to try it a second time.
Message: "No A-Chip found"
ARCOS-SP PLUS expects to work with an ARCOFI-SP whose address is A1H (A-Chip)
but does not get a response from an A-Chip.
Check the DIP switch on the hardware determining the chip address (SW 1: ON on the
SIPB 5133-SP Evaluation Board) or measure the logic level at pin 25 (AD-pin), it must be
low.
Message: "No B-Chip found"
ARCOS-SP PLUS expects to work with an ARCOFI-SP whose address is B1H (B-Chip)
but does not get a response from an B-Chip.
Check the DIP switch on the hardware determining the chip address (SW 1: OFF on the
SIPB 5133-SP Evaluation Board) or measure the logic level at pin 25 (AD-pin), it must be
high.
Semiconductor Group
277
SIPO 2163
Message:
"This version of the A-Chip is not supported" or
"This version of the B-Chip is not supported"
"This version of the ARCOFI is not supported"
ARCOS-SP PLUS has detected an PSB 2160, or PSB 2165, or an other device, but not
the PSB 2163.
Message: "No ICC/ISAC found"
The field "ICC/ISAC" is activated but ARCOS-SP PLUS cannot find an ICC or an ISAC-S
or an ISAC-P on the external hardware.
Check the configuration of the external hardware (including DIP-switches and jumpers);
does the setting in the user area match with the hardware configuration?
Message: "No EPIC/ELIC found"
The field "EPIC/ELIC" is activated but ARCOS-SP PLUS cannot find an EPIC or an ELIC
on the external hardware.
The EPIC/ELIC is part of the STUT 2000 PERCOFI Board and of the LineCard Module.
Check the configuration of the external hardware; does the setting in the user area
match with the hardware configuration?
Message: "Serial I/F Init failed at COM port"
ARCOS-SP PLUS was started with the option /V and has tried to detect one of the
hardware platforms from table 5 without success.
This message can have many reasons:
•
•
•
•
•
•
None of the boards from table 5 is connected to the COM port
The wrong cable is used (see table 5)
The wrong port number is given in the RS232.INI file
The firmware for the external hardware was not loaded before
The external hardware is not in program mode (but in loader mode instead)
The external hardware has no power supply
Message: "Unable to write the ARCOFI"
The communication with the ARCOFI-SP is disturbed.
Has anything changed that concerns the interface mode, the chip address or the
hardware configuration? Is any cable removed or loose? Does the software try to
execute a file that was created by the ARCOS-SP PLUS program for the PSB 2165? A
good idea is to quit the ARCOS-SP software and to start again with the initialization of
the hardware.
Semiconductor Group
278
SIPO 2163
Message: "Unable to read the ARCOFI"
The communication with the ARCOFI-SP is disturbed (see above)
5.2
Other Problems
Message: "Opening Coefficient-file failed"
ARCOS-SP PLUS was not able to open the file ARC63.TAB (ARC63D.TAB). Make sure
that this file is in the same directory as ARC63.EXE (ARC63D.EXE) itself.
Message: "Syntax Error or command not allowed"
An input in the command line does not match the syntax given in chapter 2.4.
Message: "Interpreter Error! File loading stopped"
A file that should be executed contains an unknown command. Probably it is not a file
that was created with ARCOS-SP PLUS for the PSB 2163.
Message: "ARCOS-SP program does not support this command"
You are using the ARCOS-SP software which is a demo-version that does not support
hardware access in any way. Only the ARCOS-SP PLUS software is able to execute the
desired command.
Message: "ARCOS-SP program does not support this window"
You are using the ARCOS-SP software which is a demo-version that does not support
hardware access in any way.
Message: "Not enough memory" or "Unable to allocate enough memory..."
The ARCOS-SP PLUS software requires at least 400 kB free conventional DOS
memory. Remove any other memory resident programs that are not necessary for
running a DOS application to get enough memory space.
Semiconductor Group
279