ICs for Communications Multipoint Switching and Conferencing Unit - Attenuation MUSAC PEB 2245 Version 1.2 Data Sheet 03.96 T2245-XV12-D1-7600 PEB 2245 Revision History: Current Version: 03.96 Previous Version: Digital Switching and Conferencing IC’s Data Book 01.94 Page (in Version 01.94) Page (in current Version) Subjects (major changes since last revision) 163 8 Version 1.2 189 34 Figure 18 (Initializing the PEB 2245 for a 4096-kHz Device Clock) corrected 201 46 Power supply current corrected 205 51 Timing corrected (tSH4 max = tCP4 – 10 ns + tCP4H) 205 51 Timing added (tSPL min = 100 ns) 216 61 Conference number description corrected – Appendix: Design sheets added Edition 03.96 This edition was realized using the software system FrameMaker. Published by Siemens AG, Bereich Halbleiter, MarketingKommunikation, Balanstraße 73, 81541 München © Siemens AG 1996. All Rights Reserved. Attention please! As far as patents or other rights of third parties are concerned, liability is only assumed for components, not for applications, processes and circuits implemented within components or assemblies. The information describes the type of component and shall not be considered as assured characteristics. Terms of delivery and rights to change design reserved. For questions on technology, delivery and prices please contact the Semiconductor Group Offices in Germany or the Siemens Companies and Representatives worldwide (see address list). Due to technical requirements components may contain dangerous substances. For information on the types in question please contact your nearest Siemens Office, Semiconductor Group. Siemens AG is an approved CECC manufacturer. Packing Please use the recycling operators known to you. We can also help you – get in touch with your nearest sales office. By agreement we will take packing material back, if it is sorted. You must bear the costs of transport. For packing material that is returned to us unsorted or which we are not obliged to accept, we shall have to invoice you for any costs incurred. Components used in life-support devices or systems must be expressly authorized for such purpose! Critical components1 of the Semiconductor Group of Siemens AG, may only be used in life-support devices or systems2 with the express written approval of the Semiconductor Group of Siemens AG. 1 A critical component is a component used in a life-support device or system whose failure can reasonably be expected to cause the failure of that life-support device or system, or to affect its safety or effectiveness of that device or system. 2 Life support devices or systems are intended (a) to be implanted in the human body, or (b) to support and/or maintain and sustain human life. If they fail, it is reasonable to assume that the health of the user may be endangered. PEB 2245 Table of Contents 1 1.1 1.2 Page 1.3 1.4 1.5 1.6 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 Pin Configuration (top view) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 Pin Definitions and Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10 Functional Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 Device Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 System Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 2 2.1 2.2 Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 Basic Functional Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 Microprocessor Interface and Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . .29 3 3.1 3.2 3.3 3.4 Operational Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 Reset State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 Initialization Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 Operation with a 4096-kHz Device Clock . . . . . . . . . . . . . . . . . . . . . . . . . . . .34 Standby Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34 4 4.1 4.2 4.3 4.4 4.5 4.6 Detailed Register Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 Mode Register (MOD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 Status Register (STA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 Conference Status Register (CST) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 Conference Mask Register (CMR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40 Indirect Access Register (IAR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40 Indirect Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .44 5 5.1 5.2 5.3 5.4 5.5 5.5.1 Electrical Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46 Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46 DC Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46 Capacitances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46 AC Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47 Microprocessor Interface Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49 Motorola Bus Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49 6 Conference Applications of the MUSACTM . . . . . . . . . . . . . . . . . . . . . . . .57 7 Package Outlines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 Semiconductor Group 3 PEB 2245 Table of Contents 8 8.1 8.2 8.3 8.4 Page Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .66 Initialization for Conferencing in a PBX . . . . . . . . . . . . . . . . . . . . . . . . . . . . .66 Programming a Conference in a PBX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .67 Programming Procedure for Switching TS’s . . . . . . . . . . . . . . . . . . . . . . . . .68 Programming Procedure for a PBX Conference . . . . . . . . . . . . . . . . . . . . . .69 IOM®, IOM®-1, IOM®-2, SICOFI®, SICOFI®-2, SICOFI®-4, SICOFI®-4µC, SLICOFI®, ARCOFI® , ARCOFI®-BA, ARCOFI®-SP, EPIC®-1, EPIC®-S, ELIC®, IPAT®-2, ITAC®, ISAC®-S, ISAC®-S TE, ISAC®-P, ISAC®-P TE, IDEC®, SICAT®, OCTAT®-P, QUAT®-S are registered trademarks of Siemens AG. MUSAC™-A, FALC™54, IWE™, SARE™, UTPT™, ASM™, ASP™ are trademarks of Siemens AG. Purchase of Siemens I2C components conveys a license under the Philips’ I2C patent to use the components in the I2C-system provided the system conforms to the I2C specifications defined by Philips. Copyright Philips 1983. Semiconductor Group 4 PEB 2245 1 Overview A Complete Family of Efficient Solutions If the issue is digital switching and conferencing, the solution is flexibility, capacity, and economy. Siemens Semiconductor offers the most economical answer to all conceivable applications in this field. Our complete family of switching network devices satisfies even the most rigorous switching demands. A Complete Family of Efficient Solutions Take our MTSC (Memory Time Switch CMOS) PEB 2045 with a switching capability of 512 incoming PCM channels to 256 outgoing PCM channels. It has the perfect size to economically build medium sized switches. The design of a non-blocking switch for 512 PCM channels is possible with a simple parallel configuration with a second MTSC. If you need a non-blocking switch for up to 256 channels, we offer a smaller version of the MTSC, the MTSS (Memory Time Switch Small) PEB 2046. And the MTSL (Memory Time Switch Large) PEB 2047, the largest in our family, is capable of switching 1024 PCM channels. Siemens also supplies the best solution for conferencing, our MUSAC (Multipoint Switching and Conferencing Unit) PEB 2245 performs the complete switching functions of the MTSC, and offers a signal processor for handling up to 64 conferencing channels in any combination. The input and output channels can also be attenuated individually to achieve best transmission quality. The MUSAC-A (Multipoint Switching and Conferencing Unit) PEB 2445 is an upward compatible device to the MTSC and MUSAC. It offers in addition the attenuation and amplification of every time slot. Pin compatible device allow simplicity in hardware and software design. To allow for more flexibility, the PCM data rate can be 2, 4, or 8 Mbit/s – configurable also for mixed use. Semiconductor Group 5 PEB 2245 The figure below shows the general architecture of a digital exchange. Figure 1 General Exchange Architecture System Background Digital exchanges put calls through by newly arranging the speech signals coded with 8-bit words (PCM time-slots). The code words are transmitted serially on PCM lines. The sampling frequency of 8 kHz produces PCM frames with a duration of 125 µs. The transmission rate on the line determines how many code words (speech channels) can be accommodated within a sampling period. With a data rate of 2048 kbit/s for example, there are 32 time-slots of 8 bits each. 4 lines with a data rate of 8192 kbit/s have a transmission capacity of 512 channels. Semiconductor Group 6 PEB 2245 An overview on the complete switching and conferencing IC-family is shown in the following table: Table 1 Complete Switching and Conferencing IC Family MTSC PEB 2045 MTSS PEB 2046 MTSL PEB 2047 MTSL 16 PEB 2047-16 MUSAC PEB 2245 MUSAC-A PEB 2445 EPIC-1 PEB 2055 EPIC-S PEB 2054 Switching capacity (time-slots) 512 × 256 256 × 256 1024 × 512 1024 × 1024 512 × 256 512 × 256 256 × 256 256 × 256 Input/output lines ‘16/8 ‘8/8 ‘16/8 ‘16/8 ‘16/8 ‘16/8 ‘8/8 SLD/IOM/ PCM ‘6/6 IOM/PCM PCM-data rate (Mbit/s) 2/4/8 + mixed mode 2 2/4/8 + mixed mode 2/4/8/16 + mixed mode 2/4/8 + mixed mode 2/4/8 + mixed mode up to 8 up to 8 Clock rate (MHz) 4.096 8.192 4.096 8.192 4.096 8.192 4.096/8.192 16.384 4.096 8.192 4.096 8.192 up to 8.192 up to 8.192 Conferencing 64 channels 64 channels Attenuation 64 channels 3/6/9 dB all channels – 4 to 12 dB yes yes PRI/T1 mode yes Fractional T1 data bundling yes yes 128-Kbit/s channel 128-Kbit/s channel µC access read read yes yes Multipoint switching yes yes Power (mW) max. consumption typ 50 50 100 170 100 100 50 50 Package P-DIP-40 P-LCC-44 P-DIP-40 P-LCC-44 P-LCC-44 P-LCC-44 P-LCC-44 P-LCC-44 P-LCC-44 P-LCC-44 1) in definition Semiconductor Group 7 03.96 Multipoint Switching and Conferencing Unit (MUSACTM) PEB 2245 Preliminary Data 1.1 CMOS IC Features Switching ● Time/space switch for 2048-, 4096- or 8192-kbit/s PCM ● ● ● ● ● ● ● ● ● ● systems Switching of up to 512 incoming PCM-channels to up to 256 outgoing PCM channels 16 input and 8 output PCM lines P-LCC-44 Different kinds of modes (2048, 4096, 8192 kbit/s or mixed mode) Configurable for primary access and standard applications Programmable clock shift with half clock step resolution for input and output in primary access configuration Configurable for a 4096- and 8192-kHz device clock Tristate function for further expansion and tandem operation Tristate control signals for external drivers in primary access configuration 2048-kHz clock output in primary access configuration Space switch mode Multipoint Switching ● Multiple independent LAN’s within one PBX ● Multiplexing of up to 64 channels ● 64-kbit/s channels Type Version Ordering Code Package PEB 2245 V1.2 Q67100-H6209 P-LCC-44 (SMD) Semiconductor Group 8 03.96 PEB 2245 Conference Mode ● ● ● ● ● ● ● ● ● ● Up to 64 conference channels in any combination Up to 21 independent conferences simultaneously (3 subscribers) Programmable attenuation (0/3/6/9 dB) on each input channel Programmable attenuation (0/3 dB) on each output channel Programmable PCM-level adaption (attenuation or amplification) of up to 64 channels Programmable noise suppression (four thresholds) Conference overflow handling Tone insertion capability A-law / µ-law compatible Compatible with all kinds of PCM-byte formats General ● ● ● ● 8-bit Motorola or Intel type µP interface Single + 5-V power supply Advanced low power CMOS technology TTL-compatible inputs/outputs 1.2 Pin Configuration (top view) Figure 2 Semiconductor Group 9 PEB 2245 1.3 Pin Definitions and Functions Pin No. Symbol Input (I) Output (O) Function 1 VSS I Ground (0 V) 6 INT OD Interrupt Request: The signal is activated when a conference overflow is detected. The microprocessor may determine the specific conference in overflow by reading the conference status register (CST). The interrupt is maskable. INT is an open drain output, thus a “wired-or” combination of interrupt request outputs of several MUSACs is possible. (A pull up resistor is necessary). 3 SP I Synchronization Pulse: The MUSAC is synchronized relative to the PCM system via this line. 4 7 9 11 13 14 15 16 17 18 19 IN1 IN5 IN9 IN13 IN14 IN15 IN10 IN11 IN6 IN7 IN2 I I I I I I I I I I I PCM-Input Ports: Serial data is received at these lines at standard TTL levels. 5 8 10 12 IN0/TSC0 IN4/TSC1 IN8/TSC2 IN12/TSC3 I/O I/O I/O I/O PCM-Input Port / Tristate Control: In standard configuration these pins are used as input lines, in primary access configuration they supply control signals for external devices. 20 IN3/DCL I/O PCM-Input Port / Data Clock: In standard configuration IN3 is the PCM input line 3, in primary access configuration it provides a 2048-kHz data clock for the synchronous interface. 21 28 A0 A1 I I Address for Direct Register Access: These pins are only active if a demultiplexed µP-interface mode is selected. 22 CS I Chip Select: A low level selects the MUSAC for a register access operation. 23 VDD I Supply Voltage: 5 V ± 5 %. Semiconductor Group 10 PEB 2245 Pin Definitions and Functions (cont’d) Pin No. Symbol Input (I) Output (O) Function 24 RD I Read: This signal indicates a read operation and is internally sampled only if CS is active. The MUSAC puts data from the selected internal register on the data bus with the falling edge of RD. RD is active low (Siemens/ Intel bus mode). DS I Data Strobe: The rising edge marks the end of a valid read or write operation (Motorola bus mode). WR I Write: This signal initiates a write operation. The WR input is internally sampled only if CS is active. In this case the MUSAC loads an internal register with data from the data bus at the rising edge of WR. WR is active low (Siemens/Intel bus mode). R/W I Read/Write: When “high”, identifies a valid µP access as a read operation. When “low”, identifies a valid µP access as a write operation (Motorola bus mode). 2 ALE I Address Latch Enable: In the Intel type multiplexed µPinterface mode a logical high on this line indicates an address of an MUSAC internal register on the external address/data bus. In the Intel type demultiplexed µPinterface mode this line is hardwired to VSS, in the demultiplexed Motorola type µP-interface mode it should be connected to VDD. 26 27 29 30 31 32 33 34 AD0 AD1 AD2 AD3 AD4 AD5 AD6 AD7 I/O I/O I/O I/O I/O I/O I/O I/O Address Data Bus: If the multiplexed address/data µPinterface bus mode is selected these pins transfer data and addresses between the µP and the MUSAC. 35 36 37 38 40 41 42 43 OUT7 OUT6 OUT5 OUT4 OUT3 OUT2 OUT1 OUT0 O O O O O O O O PCM-Output Port: Serial data is sent by these lines at standard CMOS- or TTL levels. These pins can be tristated. 25 Semiconductor Group If a demultiplexed mode is used, these bits interface with the system data bus. 11 PEB 2245 Pin Definitions and Functions (cont’d) Pin No. Symbol Input (I) Output (O) Function 39 RES I Reset: A high signal on this input forces the MUSAC into reset state. The minimum pulse length is four clock periods. 44 CLK I Clock: 4096- or 8192-kHz device clock. Semiconductor Group 12 PEB 2245 1.4 Functional Symbols Figure 3 Functional Symbol for the Standard Configuration Figure 4 Functional Symbol for the Primary Access Configuration Semiconductor Group 13 PEB 2245 1.5 Device Overview The Multipoint Switching and Conferencing Unit (MUSAC) combines a time switch unit (MTSC) and a powerful signal processor on one chip. The MUSAC enhances the capabilities of a PBX by supporting teleconferencing and multipoint data communication over voiceband channels. Digital signal processing techniques are used to implement the conferencing algorithms. Up to 64 channels of the 512 incoming PCM channels may be manipulated by the signal processor and output to any of 256 outgoing PCM channels. All functions are programmed and controlled via an 8-bit standard µP interface (Motorola or Intel type). The MUSAC is fabricated using the advanced CMOS technology from Siemens and is mounted in a P-LCC-44 package. Inputs and outputs are TTL-compatible. Figure 5 Block Diagram of the PEB 2245 Semiconductor Group 14 PEB 2245 1.6 System Integration Conferencing The MUSAC is designed to connect any of the 512 PCM-input channels to any of 256 output channels. Any input channel up to a total number of 64 can be handled in 21 independent conferences simultaneously. Any conference combination from 3 subscribers in 21 conferences up to 64 subscribers in only one conference is possible. In order to ensure an acceptable speech quality and to reduce echo and “singing” problems, the input channels can be attenuated individually by 0, – 3 dB, – 6 dB or – 9 dB and the output channels by 0 or – 3 dB; additionally, input signals below a threshold programmable to four different levels are disregarded. To lessen the risk of instability in multiparty conferences the voice signal from every second channel can be inverted so that disturbance signals in odd and even channels are subtracted from one another. If more capacity is needed, several devices can be connected. By connecting the 16 PCM-input lines in parallel to two MUSACs, a nonblocking switching matrix for 512 subscribers can be implemented: 128 input channels can be selected for up to 42 independent, simultaneous conferences. Figure 6 shows such an arrangement. Due to the tristate capability of the MUSAC larger switches with conferencing capability can be easily formed. Figure 6 Memory Time Switch 16/16 for a Non-Blocking 512-Channel Switch with Conferencing Capability Semiconductor Group 15 PEB 2245 Figure 7 shows the architecture of a primary access board with common channel signaling using four CMOS devices. Figure 7 Architecture of a Primary Access Board Multipoint Switching In a multipoint configuration the communication between different stations is done by using a common media. In a PBX system this can be achieved by connecting all stations to one (or more) time-slots and transmitting the information back. Multipoint switching is a special form of conferencing for data communication. In contrast to audio conferences terminals broadcast data to the MUSAC which are only “or-connected”. That is, at each bit time, the “conference sum” is “1” if the input of one or more terminals is “1”; otherwise, the result is “0”. A simple example of such a system using Siemens VLSI switching devices is shown in figure 8. ISDN subscribers are connected via line cards and PCM highways to a multipoint switching matrix. The data from different terminals are summed up in the multipoint switching matrix and transmitted back to all stations. The switching matrix is build by using just one MUSAC. Every combination of subscribers may be switched to the same transport media (time-slot), in this way enabling a number of powerful multipoint communication systems. Semiconductor Group 16 PEB 2245 Figure 8 Multipoint System Configuration for ISDN Subscribers Semiconductor Group 17 PEB 2245 In order to establish a multipoint-connection with more than 64 terminals, you can form a multistage arrangement, as shown in figure 9. Figure 9 Multistage Arrangement Semiconductor Group 18 PEB 2245 2 Functional Description Figure 10 Detailed Block Diagram of the PEB 2245 2.1 Basic Functional Principles The MUSAC is a memory time switch device for a PCM PBX system, offering a variety of additional features like multipoint switching, conference calls, programmable noise suppression and attenuation. The MUSAC works either in standard configuration for usual switching applications or in the primary access configuration, where it realizes, together with the PEB 2035 (ACFA) and the PEB 2235 (IPAT), the system interface for up to four primary multiplex access lines. In both configurations the conference and multipoint switching capability can be used. Semiconductor Group 19 PEB 2245 The block diagram is shown in figure 10. The MUSAC is designed to connect any of 512 PCM-input channels to any of 256 output channels. Any input channel up to a total number of 64 can be handled in 21 independent conferences simultaneously. Any conference combination from 3 subscribers in 21 conferences up to 64 subscribers in only one conference is possible. Not more than 8 subscribers should be connected to a single conference, however, in order to ensure an acceptable speech quality. It can be improved by selecting an additional attenuation and activating the noise suppression: The input channels can be attenuated by 0, – 3 dB, – 6 dB or – 9 dB and the output channels by 0 or – 3 dB. Input signals below a threshold programmable to four different levels are disregarded. The input information of a complete frame is stored in the on-chip 4-Kbit Speech Memory (SM). The incoming 512 channels of 8 bits each are written in sequence into fixed positions in the SM with a repetition rate of 8 kHz. Additionally, in the second half of the frame the 64 conference output channels of 8 bit each are written into the SM. The memory access is normally controlled by the input counter in the timing control block when writing into the SM but by the conference unit when writing the conference output channels. The read access is independent of the write access, so that both input and conference output channels can be read at any time. For outputting, the Connection Memory (CM) is read in sequence. Each location in the CM points to a location in the Speech Memory. The byte in this SM location is read into the current output timeslot. The read access of the CM is controlled by the output counter also contained in the timing control block. In addition, in the first half of the frame the input channels connected to a conference are read in sequence by the Conference Unit (CU). All connections are set up by an external controller which programs the Connection Memory (CM) and the Conference Control Memory (CCM) using the microprocessor interface. The CM address corresponds to one particular output time-slot and line number. The contents of this CM-location points to a particular input time-slot and line number in the transparent mode. In the conference mode or multipoint switching mode it contains the conference address and points to a conference output location in the SM instead. The same conference address is used to access the CCM. The parameters stored in the CCM include the input time-slot and line number, the associated conference number as well as the noise suppression thresholds and the attenuation levels. The conference number defines a unique location in the Conference Sum Memory (CSM) used to store the accumulated samples for each conference. The Conference Sum Memory is alternately loaded in the first half of the frame and unloaded in the following second half. In the first half the input samples are processed to implement the noise suppression, the expansion according to the European A-law or the US µ-law and the attenuation function. The Data Memory (DM) buffers these samples for output processing. The CSM is used to accumulate these samples and store the resulting sum. During output processing the input sample is retrieved from the Data Memory and the appropriate sum from the Conference Sum Memory for subtraction, so that the channel output signal contains the contribution of all the other channels in the conference except its own. After output attenuation and PCM compression, the data are written in the Speech Memory for output switching. If one result of the subtractions exceeds the full scale value, a saturation appears and the MUSAC signals this conference overflow condition by an interrupt. The conference number of the conference in overflow is buffered in the Conference Status Register (CST) which can be retrieved by the external controller. Semiconductor Group 20 PEB 2245 A tone to be inserted into a conference is handled as an additional conference subscriber using any input PCM channel (access to CCM) but without assigning an output time-slot (no access to CM). Multipoint switching is a special form of conferencing for data communication. In the multipoint switching mode several terminals are connected together. Normally only one should transmit at a time; its signal is distributed to the other terminals. For collision detection purposes all input signals are summed up to construct the output signal. In contrast to audio conferences terminals broadcast data to the MUSAC which are only “or-connected”. That is, at each bit time, the “conference sum” is “1” if the input of one or more terminals is “1”; otherwise, the result is “0”. The data memory, the subtractor, the linearization and attenuation are of no use in this mode. The general procedure is the same as for conferencing. The chip architecture makes it possible to decrease the delay between incoming and outgoing PCM channels. The processed input samples are transmitted either in the same frame or in the next frame at the latest. Definitions ● The PEB 2245 works with either an 8192-kHz clock or a 4096-kHz clock. Henceforth, the respective clock periods are referred to as tCP8 and tCP4. ● The bits of a time-slot are numbered 0 through 7. Bit 0 (MSB) of a time-slot is the first bit to be received or transmitted by the MUSAC, bit 7 (LSB) the last. Preparation of the Input Data (Input Buffer) The PEB 2245 works in 2048-, 4096- or 8192-kbit/s PCM systems. The frame frequency is 8000 Hz in all 3 types of systems. Therefore a frame consists of 32, 64 or 128 time-slots of 1 byte each, respectively. In order to fill the speech memory, which has a fixed capacity of 512 channels, either 16-, 8- or 4 input lines are necessary, respectively. Thus, in 4- and 8-MHz systems only some of the 16 input lines can be used. Moreover, the PEB 2245 can also work with two different input data rates simultaneously. In this case some of the PCM-input lines operate at one data rate, while others operate at another. Table 2 states how many input lines are operating at the different data rates for all possible input data rate combinations. In the following they will be referred to as input modes. The input mode the PEB 2245 is actually working in has to be programmed into the mode register, bits MI1, MI0, MO1, MO0. In chapter 4.1 you will find a complete description which input line is connected to which system, for each of the input modes. Semiconductor Group 21 PEB 2245 Table 2 Possible Input Modes Input Modes Type 16 × 2048 kbit/s Single mode 8 × 4096 kbit/s Single mode 4 × 8192 kbit/s Single mode 2 × 8192 + 8 × 2048 kbit/s Mixed mode 4 × 4096 + 8 × 2048 kbit/s Mixed mode The PEB 2245 runs with either a 4096- or a 8192-kHz device clock as selected with CFR:CPS. Data rates and clock frequencies may be combined freely. However, processing 8192-kbit/s data, an 8192-kHz clock must be supplied. The preparation of the input data according to the selected input mode is made in the input buffer. It converts the serial data of a time-slot to parallel form. In standard configuration time-slot 0 begins with the rising edge of the SP pulse as shown in upper half of figure 11 denoted CSR:(0000XXXX). As can be seen there the beginning of a input time-slot is defined such, that the input lines have settled to a stable value, when the datum is actually sampled. 4096- and 8192-kbit/s data is sampled in the middle of the bit period at the falling edge of the respective data clock. 2048-kbit/s data is sampled after 3/4 of the according bit period, i.e. with the rising edge of the 4th 8192-kHz clock cycle or the falling edge of the 2nd 4096-kHz clock cycle of the considered bit period. In the primary access configuration a different timing scheme may apply to the odd (physical) input lines. They are affected by the content of the clock shift register (CSR), which can be programmed via the µP interface (see paragraph 2.2). The clock shift register holds the information, how the frame structure is shifted in the primary access configuration. Its content defaults to 00H after power up and is also set to this value, whenever the standard configuration is selected. Semiconductor Group 22 PEB 2245 Figure 11 Latching Instant for Input Data Semiconductor Group 23 PEB 2245 The four most significant bits of the clock shift register are of interest for the input lines. They only affect the odd input lines (see section Clock Shift Register): The frame structure can be advanced by the number of bit periods programmed to the RS2, RS1 and RS0 bits of the CSR. For example, programming the CSR with (1100XXXX) a new frame starts 6-bit periods before the rising edge of the SP pulse. Selecting RRE to logical 1 the frame is delayed by half a bit period (see figure 11). The data is then sampled in the middle of the respective bit period for all data rates. The last line of figure 11 shows the sampling instants for the CSR entry (1001XXXX). Then the input frame is advanced by 4-bit periods and delayed by a half resulting in a 3 1/2 clock period advancement of the input frame. For further examples refer to figure 19. Thus the frame structure may be selected to begin at any 1/2-bit period value between a resulting advancement of 7-bit periods and a resulting delay of 1/2 a bit period. Setting CSR = 0XH the same timing conditions apply to even and odd inputs. Then all system interface inputs are processed in the same way they are in the standard configuration. Output Buffer The output buffer rearranges the data read from the speech memory. It basically converts the parallel data to serial data. Depending on the validity bit the output buffer outputs the data or switches the line to high impedance. The most significant bits of the 256 words in the connection memory are interpreted as validity bits for the 256 possible output channels: A logical 0 enables the programmed connection, a logical 1 tristates the output. The mode register (MOD) bits MI1, MI0, MO1 and MO0 control this process. The possible output modes are listed in table 3. Table 3 Possible Output Modes Output Modes Type 8 × 2048 kbit/s Single mode 4 × 4096 kbit/s Single mode 2 × 8192 kbit/s Single mode 1 × 8192 + 4 × 2048 kbit/s Mixed mode 2 × 4096 + 4 × 2048 kbit/s Mixed mode Figure 12 shows when the single bits are output. In standard configuration they are clocked off at the rising clock edge at the beginning of the considered bit period. Time-slot 0 starts two tCP8 before the falling edge of the SP pulse. In primary access configuration the even output lines are affected by the XS2, XS1, XS0 and XFE entries in the clock shift register. The output frame is synchronized with the rising edge of the SP signal. Semiconductor Group 24 PEB 2245 Assuming a CSR entry X0H the output frame starts with the rising edge of the SP pulse. Programming the XS2, XS1 and XS0 bits with a value deviating from binary 000 the output frame is delayed by 8D-(XS2, XS1, XS0)B bit periods. E.g., a CSR entry of (XXXX0010) delays the output frame by 7-bit periods relative to the rising SP-pulse edge. Programming CSR:(XXXXXXX1) the output frame is delayed by another half a device clock period. In figure 12 the outputting instants are shown for a device clock of 4096 and 8192 kHz and a CSR:(XXXX0001). The last line in figure 12 shows an even 8192-kbit/s output line for the CSR entry (XXXX1101) and an 8192-kHz device clock. The output frame is delayed by 2 1/2-bit periods. For further examples refer to figure 19. If the CSR is programmed such that XS2 is identical to RS2, XS1 to RS1, XS0 to RS0 and RRE to XFE the time-slot boundaries of input and output coincide. Programming XS2, XS1, XS0 as well as RS2, RS1, RS0 to logical 0 input and output time-slots coincide. Otherwise the system interface output frame starts one time-slot after the system interface input. This can be seen comparing for example the lines 0100XXXX and XXXX0100 in figure 19. Semiconductor Group 25 PEB 2245 SP Primary Access Configuration SP Standard Configuration 2 t CP8 CLK 8192 kHz CLK 4096 kHz CSR: (XXXX 0000) 5 6 7 bit0 1 2 3 5 7 0 1 2 CSR: (XXXX 0001) 8192 kHz clock 1 3 0 2 4 1 6 7 bit0 1 2 - Slot 127 Time 6 3 4 5 6 7 0 - Slot 0 Time 7 0 1 5 4096 kbit/s Data Rate 2 2048 kbit/s Data Rate 1 8192 kbit/s Data Rate 2 2 3 4096 kbit/s Data Rate 4 - Slot 0 Time 7 0 1 - Slot 31 Time 2 2048 kbit/s Data Rate Time- Slot 0 6 7 0 1 2 3 4 7 0 2048 kbit/s Data Rate 1 - Slot 31 Time Time- Slot 0 CSR: (XXXX 1101) 5 4096 kbit/s Data Rate Time- Slot 0 - Slot 63 Time 6 7 bit0 1 2 - Slot 127 Time 3 4 - Slot 0 Time Figure 12 Clocking Off Instant of Output Data Semiconductor Group 8192 kbit/s Data Rate Time- Slot 1 - Slot 63 Time 4 3 Time- Slot 0 - Slot 31 Time CSR: (XXXX 0001) 4096 kHz clock 0 - Slot 0 Time 7 5 7 - Slot 1 Time Time - Slot 63 4 6 - Slot 0 Time - Slot 127 Time 6 4 26 5 6 7 0 8192 kbit/s Data Rate ITD03747 PEB 2245 Configuration Type The MUSAC works either in the standard configuration for usual switching applications or in the primary access configuration. In these both configurations the conference and multipoint switching capability can be used. Standard Configuration A logical 1 in the CFS bit of the configuration register sets the PEB 2245 in standard mode (default after power up). All modes from table 7 can be used. It has to be ensured that the data rate is not higher than the selected device clock (4096 or 8192 kHz). In this application 512 channels per frame are written into the speech memory. Each one of them can be connected to any output channel. According to table 8 and table 10 and depending on the selected mode the least significant bits of the connection memory address and data contain the logical pin numbers, the most significant bits the time-slot number of the output and input channels. The following example explains the programming sequence. Time-slot 7 of the incoming 8192-kbit/s input line IN 14 shall be connected to time-slot 6 of the output line OUT 5 of an 2048-kbit/s system. According to table 8 in 8192-kbit/s systems the input line IN 14 is the logical input line 2. Output line number and logical output number are identical to one another. Therefore the following byte sequence on the address data bus has to be used to program the CM properly (see table 10). 00100000 00011110 00110101 (Control Byte) (Data Byte) (Address Byte) The frame, for all input channels, starts with the rising edge of the SP signal. The frame for all output channels begins two tCP8 (with 8192-kHz device clock) or one tCP4 period (4096-kHz device clock) before the falling SP edge. The period of time between the rising and falling edge of the SP pulse should be tSPH = = (2 + N × 4) tCP8 (1 + N × 2) tCP4 (0 ≤ N ≤ 255) N is an user defined integer. By varying N, tSPH can be varied in 2048-kHz clock period steps. For an example using N = 2 refer to figure 13. Semiconductor Group 27 PEB 2245 Figure 13 SYP Duration for N = 2 Primary Access Configuration A logical 0 in the CFS bit of the configuration register selects the PEB 2245 for primary access applications. In this case the MUSAC is an interface device connecting a standard PCM interface (system interface) with another PCM-interface e.g. an intermediate interface for connections to primary loops (synchronous interface). For both a serial interface is provided. The synchronous 2048-kbit/s interface consists of four input and four output lines with a bit rate of 2048-kbit/s. This interface can be used to connect the PEB 2245 to up to four primary trunk lines via coding/decoding devices with frame alignment function (e.g. PEB 2035 ACFA) and line transceivers with clock and data recovery (e.g. PEB 2235 IPAT) and to signalling processors (e.g. the SAB 82520 HSCC). The system interface is not confined to one data rate but can operate at the full choice of the PEB 2245 data rates: 2048, 4096 and 8192 kbit/s. A clock shift in a range of 7 1/2 clock steps with half clock step resolution may be programmed independently for inputs and outputs. The frame for all input- and output lines starts with the rising edge of the SP signal. In the primary access mode the signals TSC0, TSC1, TSC2 and TSC3 indicate when the associated system interface output is valid. The signal DCL supplies a 2-MHz clock which can be used for other devices at the synchronous interface, e.g. the High Level Serial Communication Controller HSCC (SAB 82520). In the primary access configuration only those modes which support at least 4 input and 4 output lines at 2048-kbit/s can be used. These are the modes MI1, MI0, MO1, MO0 = 0H, AH, FH (see table 7). Programming the CM in the primary access configuration is described in tables 8, 11 and 12. The least significant 2 bits of the data byte and the least significant bit of the address byte determine the type of interface, the more significant bits define the logical line number and time-slot number. According to figure 14 in the primary access configuration the connection memory is usually programmed to switch the system and synchronous interface inputs to the synchronous and system interface outputs, respectively. However, it is also possible to connect the system interface inputs to the system interface outputs as well as the synchronous interface inputs to the synchronous interface outputs. This connection possibility allows for test loops at the system and the synchronous interfaces. Semiconductor Group 28 PEB 2245 Figure 14 Connection Choices in the Primary Access Configuration 2.2 Microprocessor Interface and Registers The MUSAC is programmed via the µP interface. It consists of the address data bus AD7 … AD0, the address bits A1 … A0, the Write (WR), the Read (RD), the Address Latch Enable (ALE), the Interrupt (INT) and the Chip Select (CS) signal, as shown in figure 15. Figure 15 The MUSACTM Controlled by a Intel Microprocessor The standard 8-bit µP interface can communicate with Intel multiplexed/demultiplexed microprocessors as well as with Motorola demultiplexed processors. It gives access to the internal registers and to the control memories (Connection Memory, Conference Control Memory). Semiconductor Group 29 PEB 2245 Table 4 µP-Interface Functions ALE Type of µP-Interface Bus Structure Pin 24 Pin 25 Package Fixed to VDD Fixed to ground Switching Motorola Intel Intel demultiplexed demultiplexed multiplexed DS RD RD R/W WR WR P-LCC P-LCC P-LCC In the multiplexed µP-interface mode A0, A1 have to be fixed to logical 0. For a demultiplexed µP interface the address bits A1 and A0 are needed for addressing a register. For the µP-interface timing pleasure refer to chapter 5.5. Five directly addressable registers are provided: ● ● ● ● ● Mode register (MOD) Status register (STA) Conference Status register (CST) Conference Mask Register (CMR) Indirect Access Register (IAR) Two other registers and the control memories are accessed by a simple three byte indirect access method: ● ● ● ● Configuration Register (CFR) Clock Shift Register (CSR) Connection Memory (CM) Conference Control Memory (CCM) The status register (STA) and the conference status register (CST) are read-only-registers, the conference mask register (CMR) is a write-only-register; MOD, CFR, CSR, IAR and CM or CCM can be read or written. An indirect access scheme is used to access the CFR, CSR, CM or CCM using the indirect access register (IAR). The following direct registers may be accessed: Table 5 Addressing the Direct Registers Address Write Operation Read Operation Demultiplexed Mode A (1:0) Multiplexed Mode AD (7:0) 0H 0H MOD MOD 1H 2H – STA 2H 4H CMR CST 3H 6H IAR IAR Semiconductor Group 30 PEB 2245 Indirect access to the CFR, CSR, CM or CCM: An indirect access is performed by reading/writing three consecutive bytes (first byte = control byte, second byte = data byte, third byte = address byte) to/from IAR. Bit 7 Bit 0 K2 K1 K0 0 0 DI0 D9 D8 Control Byte D7 D6 D5 D4 D3 D2 D1 D0 Data Byte IA7 IA6 IA5 IA4 IA3 IA2 IA1 IA0 Address Byte The bits K2 … K0 determine whether the CFR, the CSR, the CM or the CCM shall be accessed, whether a write or read operation shall be performed and whether the first or the second memory access shall be executed. (To describe a conference two accesses to the CCM are necessary). The bits D7 … D0 contain the information which shall be written into the control memories or the indirect registers. The address byte indicates which one of the indirect registers shall be accessed or in which memory location the data shall be written. An exact definition is given in chapter Indirect Access Register. Before an indirect access is started, the Z- and B bits of the status register must be 0. With the first instruction the Z-bit is set (see chapter Status Register). After the third instruction the MUSAC accesses the memory location. This access requires maximally 900 ns. After the access is finished the Z bit is reset. Figure 16 a) illustrates a write operation on the IAR. It is possible to read or write the direct access registers while an indirect access is in progress. Thus a register may be read in the time intervals that separate the three sequential indirect access instructions. Also, the current indirect access may be aborted by setting the MOD:RI. One indirect register access has to be completed before the next one can be started. To read the indirect registers or the CM two sequences of three instructions each have to be programmed. In the first sequence the MUSAC is instructed which register of CM address to read. The data transferred to the PEB 2245 in this first sequence is of no significance. With the first write instruction STA:Z is set. After the first 3 instructions the MUSAC needs 900 ns to write the result to the IAR. The status register bit Z is reset after maximally 900 ns. Then 3 read operations follow. Again, STA:Z is set with the first read instruction. The 3 instructions read 3 bytes from the IAR. Figure 16 b) shows this procedure. After the third read operation the PEB 2245 needs another 900 ns to reset the indirect access mechanism and the Z bit in the status register. In CCM read accesses three sequences of three instructions (two write sequences and one read sequence; see figure 16 c)) are necessary. Semiconductor Group 31 PEB 2245 Figure 16 Timing Diagrams of IAR Semiconductor Group 32 PEB 2245 3 Operational Description 3.1 Reset State After a hardware reset (RES) the MUSAC is set to its initial state. The MOD- and CFR-register bits are all set to logical 1; the CSR, CST and CMR register bits are set to logical 0. The STA register B bit is undefined, the Z bit contains logical 0. 3.2 Initialization Procedure After reset a few internal signals and clocks need to be initialized. This is done with the initialization sequence. To give all signals and clocks a defined value only 4 SP pulses are necessary. The SP pulses may be of any length allowed in normal application, the time interval between the two SP pulses may be of any length down to 250 ns. With all signals being defined, the CM needs to be reset. To do that a logical 0 is written into MOD:RC. STA:B is set. The resulting CM reset is finished after max. 250 µs and is indicated by the status register B bit being logical 0. Changing the pulse shaping factor N during CM-reset may result in a CM-reset time longer than 250 µs. To prepare the MUSAC for programming the CM and CCM, the RI bit in the mode register must be reset. Note that one mode register access can serve to reset both RC- and RI bits as well as configuring to chip (i.e. selecting operating mode etc.). Figure 17 Initializing the PEB 2245 for a 8192-kHz Device Clock Semiconductor Group 33 PEB 2245 Figure 18 Initializing the PEB 2245 for a 4096-kHz Device Clock 3.3 Operation with a 4096-kHz Device Clock In order for the MUSAC to operate with a 4096-kHz device clock the CPS bit in the CFR register needs to be reset. This has to be done before the CM reset and needs 1.8 µs. For a flow chart of this process refer to figure 18. 3.4 Standby Mode With MOD:SB being logical 1 the MUSAC works as a backup device in redundant systems. It can be accessed via the µP interface and works internally like an active device. However, the outputs are high impedance. If the SB bit is reset, the outputs are switched to low impedance for the programmed active channels and this MUSAC can take over from another device which has been recognized as being faulty. Semiconductor Group 34 PEB 2245 4 Detailed Register Description The following registers may be accessed: Table 6 Addressing the Direct Registers Address Write Operation Read Operation Demultiplexed Mode A (1:0) Multiplexed Mode AD (7:0) 0H 0H MOD MOD 1H 2H – STA 2H 4H CMR CST 3H 6H IAR IAR The chapters in this section cover the registers in detail. 4.1 Mode Register (MOD) Access in the multiplexed µP-interface mode: Read/write, address: 0H Access in the demultiplexed µP-interface mode: Read/write, address: 0H Reset value: BFH AD7 RC AD0 0 RI SB MI1 MI0 MO1 MO0 RC Reset Connection memory; writing a zero to this bit causes the complete connection memory to be overwritten with 200H (tristate). During this time STA:B is set. The maximum time for resetting is 250 µs. RI Reset Indirect access mechanism; setting this bit resets the indirect access mechanism. RI has to be cleared before writing/reading IAR after reset. SB Stand By; by selecting SB = 1 all PCM outputs are tristated. The connection memory works normally. The MUSAC can be activated immediately by resetting SB. MI1/0 MO1/0 Input/Output operation Mode; these bits define the bit rate of the input and output lines. The bitrates are given in table 7, the corresponding pin functions in table 8 (standard configuration). Semiconductor Group 35 PEB 2245 Table 7 Input/Output Operating Modes MI1 MI0 MO1 MO0 Input Mode 0 0 0 0 0 0 1 1 1 0 0 1 1 1 0 0 0 1 1 1 0 0 0 0 1 1 0 1 0 0 1 0 0 1 0 0 1 1 1 1 1 0 0 1 0 0 1 0 0 1 0 1 1 1 1 1 16 × 2 16 × 2 16 × 2 4×8 4×8 4×8 2×8/8×2 2×8/8×2 2×8/8×2 8×4 4×8 4×4/8×2 8×4 16 × 8 1 1 1 1 0 1 0 0 unused unused Output Mode Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s 8×2 2×8 4×2/1×8 8×2 2×8 4×2/1×8 8×2 2×8 4×2/1×8 4×4 4×4 4×2/2×4 2×8 2×8 Mbit/s** Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s Mbit/s** Mbit/s Mbit/s Mbit/s** Mbit/s Mbit/s* unused unused * for space switch application only; the conference or multipoint switching capability cannot be used in this operating mode ** can also be used for primary access configuration Note: In the mixed modes the first bit rate refers to the odd line numbers, the second one to the even line numbers. Semiconductor Group 36 PEB 2245 Table 8 Input and Output Pin Arrangement for the Standard Configuration Input Pin Arrangement Pin No. 16 × 8 Mbit/s 16 × 2 Mbit/s 4 5 7 8 9 10 11 12 13 14 15 16 17 18 19 20 IN1 IN0 IN5 IN4 IN9 IN8 IN13 IN12 IN14 IN15 IN10 IN11 IN6 IN7 IN2 IN3 Note: 4×8 Mbit/s 8×2+2×8 Mbit/s 8×2+4×4 Mbit/s IN0 IN0 IN4 IN4 IN8 IN1 IN0 IN2 IN3 8×4 Mbit/s IN1 IN12 IN14 IN3 IN10 IN1 IN0 IN5 IN4 IN6 IN7 IN2 IN3 IN1 IN8 IN5 IN12 IN14 IN7 IN10 IN3 IN6 IN6 IN2 IN2 The input line numbers shown are the logical line numbers to be used for programming the connection memory and the conference control memory. In the case of 16 input lines the logical line numbers are identical to the pin names. Output Pin Arrangement Pin No. 8×2 Mbit/s 35 36 37 38 40 41 42 43 OUT7 OUT6 OUT5 OUT4 OUT3 OUT2 OUT1 OUT0 Note: 2×8 Mbit/s 4×2+1×8 Mbit/s 4×2+2×4 Mbit/s OUT7 OUT7 OUT5 OUT5 OUT3 OUT1 OUT0 4×4 Mbit/s OUT1 OUT0 OUT3 OUT2 OUT1 OUT0 OUT3 OUT2 OUT1 The logical output line numbers shown above are identical to the pin names. Semiconductor Group 37 OUT0 PEB 2245 Table 9 Input, Output and Tristate Pin Arrangement for the Primary Access Configuration Pin No. System Interface Mode Pin Name P-LCC 2 MHz 4 MHz 8 MHz TSC0 TSC1 TSC2 TSC3 5 8 10 12 TSC0 TSC1 TSC2 TSC3 TSC0 TSC1 TSC0 System interface tristate control signals, clock shift programmable OUT0 OUT2 OUT4 OUT6 43 41 38 36 OUT0 OUT1 OUT2 OUT3 OUT0 OUT1 OUT0 System interface outputs clock shift programmable IN13 IN9 IN5 IN1 11 9 7 4 IN3 IN2 IN1 IN0 IN1 IN0 IN0 System interface inputs, clock shift programmable OUT1 OUT3 OUT5 OUT7 42 40 37 35 OUT0 OUT1 OUT2 OUT3 OUT0 OUT1 OUT2 OUT3 OUT0 OUT1 OUT2 OUT3 Synchronous 2-MHz interface outputs IN14 IN10 IN6 IN2 13 15 17 19 IN3 IN2 IN1 IN0 IN3 IN2 IN1 IN0 IN3 IN2 IN1 IN0 Synchronous 2-MHz interface inputs 0000 1111 1010 MI1, MI0, MO1, MO0 Mode Note: The input, output and tristate control line numbers shown in the center columns of this table are logical line numbers. The corresponding pin names are listed in the left most column. Semiconductor Group 38 PEB 2245 4.2 Status Register (STA) Access in the multiplexed µP-interface mode: Read, address: 2H Access in the demultiplexed µP-interface mode: Read, address: 1H AD7 AD0 B Z X X X VN2 VN1 VN0 X don’t care B Busy: The chip is busy resetting the connection memory (B = 1). B is undefined after power up and logical 0 after the device initialization. The three byte indirect access register is not accessible. Note: The maximum time for resetting is 250 µs. Z Incomplete instruction; a three byte indirect instruction is not completed (Z = 1). Z is 0 after power up. Note: Z is reset and the indirect access is cancelled by setting MOD:RI or resetting MOD:RC. VN (2:0) Version Number according to the table below: VN2 VN1 VN0 Device Versions 0 0 0 A1 (V1.2) 4.3 Conference Status Register (CST) Access in the multiplexed µP-interface mode: Read, address: 4H Access in the demultiplexed µP-interface mode: Read, address: 2H Reset value: 00H AD7 X AD0 IR COV CN4 CN3 CN2 CN1 CN0 X don’t care IR Initialization Request. The connection memory and the conference control memory may have lost data (IR = 1). The IR bit is set after power failure or inappropriate clocking and is reset by reading CST. COV Conference Overflow (overflow = > logical 1) CN4 … CN0 Conference Number of the conference in overflow Semiconductor Group 39 PEB 2245 4.4 Conference Mask Register (CMR) Access in the multiplexed µP-interface mode: Read/Write, address: 4H Access in the demultiplexed µP-interface mode: Read/Write, address: 2H Reset value: 00H AD7 AD0 0 IR COV 0 0 0 0 0 A logical 1 disables the corresponding interrupt. IR Initialization Request mask; the initialization request is masked (IR = 1) COV Conference Overflow mask; the conference overflow is masked (COV = 1) 4.5 Indirect Access Register (IAR) Access in the multiplexed µP-interface mode: Read/Write, address: 6H Access in the demultiplexed µP-interface mode: Read/Write, address: 3H An indirect access is performed by reading/writing three consecutive bytes (first byte = control byte, second byte = data byte, third byte = address byte) to/from IAR. Bit 7 Bit 0 K2 K1 K0 0 0 DI0 D9 D8 Control Byte D7 D6 D5 D4 D3 D2 D1 D0 Data Byte IA7 IA6 IA5 IA4 IA3 IA2 IA1 IA0 Address Byte K2 … K0 control the indirect access according to the following table: K2 K1 K0 Type of Access 0 0 1 Write CM 0 0 0 Read CM 0 1 1 Write CCM: first access 0 1 0 Read CCM: first access 1 0 1 Write CCM: second access 1 0 0 Read CCM: second access 1 1 1 Write indirect register 1 1 0 Read indirect register Semiconductor Group 40 PEB 2245 Access to CM DIO = 0 Transparent Switching (i.e. the MUSAC works exactly like a MTSC) D9 Validity bit: A logical 0 enables the programmed connection, a logical 1 tristates the outputs D8 … D0 Logical line and time-slot number of the inputs IA7… IA0 Logical line and time-slot number of the outputs D9 – D0 Is written to the CM address IA7 – IA0 DIO = 1 Conference Switching or Multipoint Switching Mode D9 Validity bit: A logical 0 enables the programmed connection, a logical 1 tristates the outputs D8 … D6 Logical 0 D5 … D0 Conference control address IA7 … IA0 Logical line and time-slot number of the outputs D9 – D0 Is written to the CM address IA7 – IA0. D5 – D0 contain the address which points to the appropriate CCM location. D8 – D0 and IA7 – IA0 contain the information for the logical line and time-slot numbers of the programmed connection, D8 – D0 for the inputs, IA7 – IA0 for the outputs. Table 10 shows the programming of these bits for standard configuration, table 11 and 12 for the primary access configuration. Table 10 Time-Slot and Line Programming for Standard Configuration Standard configuration, all modes except space switch mode 2-Mbit/s input lines Bit Bit Bit D3 D8 D9 to to D0 D4 Logical line number Time-slot number Validity bit 4-Mbit/s input lines Bit Bit Bit D2 D8 D9 to to D0 D3 Logical line number Time-slot number Validity bit 8-Mbit/s input lines Bit Bit Bit D1 D8 D9 to to D0 D2 Logical line number Time-slot number Validity bit 2-Mbit/s output lines Bit Bit IA2 IA7 to to IA0 IA3 Line number Time-slot number 4-Mbit/s output lines Bit Bit IA1 IA7 to to IA0 IA2 Line number Time-slot number 8-Mbit/s output lines Bit Bit IA0 IA7 IA1 Line number Time-slot number Semiconductor Group to 41 PEB 2245 Table 11 Time-Slot and Line Programming for the Primary Access Configuration 2-Mbit/s input lines Bit Bit Bit Bit D1 D3 D8 D9 to to to D0 D2 D4 Interface select in Line number Time-slot number Validity bit 4-Mbit/s input lines Bit Bit Bit Bit D1 D2 D8 D9 to D0 to D3 Fixed to 01 (system interface) Line number Time-slot number Validity bit 8-Mbit/s input lines Bit Bit Bit D1 D8 D9 to to D0 D2 2-Mbit/s output lines Bit Bit Bit IA0 IA2 IA7 to to IA1 IA3 Interface select out Line number Time-slot number Bit Bit Bit IA0 IA1 IA7 to IA2 Fixed to 0 (system interface) Line number Time-slot number Bit Bit IA0 IA7 to IA1 Fixed to 0 (system interface) Time-slot number 4-Mbit/s output lines 8-Mbit/s output lines Fixed to 01 (system interface) Time-slot number Validity bit The interface select bits have to be programmed as shown in the following table: Table 12 Interface Selection Bits System Interface Synchronous 2-MHz Interface Input Lines 01 10 Output Lines 0 1 Access 1 to CCM DI0 Logical 0 D9 Inversion bit: In a multiparty conference there is some risk of instability due to reflections at the hybrid. If these reflections are not canceled, they will be summed up in the conference sum and will be transmitted to the subscriber, where again they could be reflected. In very big conferences (>> 4 subscribers) this behaviour could result in an instability. To avoid this the PEB 2445 has the ability to invert every second conference channel, which has no audible influence on the speech quality. By this the noise due to reflections is compensated to a high degree. The feature can also be applied to reduce noise due to line impedance mismatch. D8 … D0 Logical line and time-slot number of the inputs (see table 10 to 12) IA7/IA6 Logical 0 IA5 … IA0 Conference control address Semiconductor Group 42 PEB 2245 Access 2 to CCM DI0 Logical 0 D9/D8 Noise suppression threshold D7/D6 Input attenuation level D5 Output attenuation level Note: D7, D6, D5 are only relevant to conference mode; in the multipoint switching mode these bits must be logical 0. D4 … D0 Conference number: 21 independent simultaneous conferences are possible. By using the conference number 1FH an attenuation or noise suppression can be inserted in a channel without conferencing. IA7/IA6 Logical 0 IA5 … IA0 Conference control address D9 D8 Noise Suppression Threshold 0 0 no noise suppression 0 1 fifth step, first segment 1 0 ninth step, first segment 1 1 sixteenth step, first segment D7 D6 Input Attenuation Level 0 0 0 dB 0 1 3 dB 1 0 6 dB 1 1 9 dB D5 Output Attenuation Level 0 0 dB 1 3 dB Please note: The sequence of programming (access 1, access 2) is important. Semiconductor Group 43 PEB 2245 4.6 Indirect Registers Configuration Register (CFR) Access: Read or write at address FEH Reset value: FFH AD7 AD0 1 1 CUA2 CUA1 CUA0 FS CFS CPS CPS Clock Period Select: Device clock is set to 8192 kHz (logical 1) or 4096 kHz (logical 0). CFS Configuration Select: The MUSAC works either in the primary access configuration (CFS = 0) or in the standard configuration (CFS = 1). FS Function Select: FS = 0: Multipoint switching FS = 1: Conferencing CUA0 …CUA2 PCM encoding law and PCM-byte format CUA0 Encoding Law 1 A-law 0 µ-law CUA2 CUA1 PCM-Byte Format 0 0 no bits inverted 0 1 even bits inverted 1 0 odd bits inverted 1 1 all bits inverted Clock Shift Register (CSR) Access: Read or write at address FFH Reset value: 00H AD7 RS2 RS2 … RS0 AD0 RS1 RS0 RRE XS2 XS1 XS0 XFE Receive clock Shift, bits 2 – 0. The received data stream is shifted in bit period steps. Semiconductor Group 44 PEB 2245 RRE Receive with Rising Edge. The data is sampled with the falling (RRE = 0) or rising edge (RRE = 1) of the data equivalent clock. XS0 … XS2 Transmit clock Shift, bits 2 – 0. The transmitted data stream is shifted. XFE Transmit with Falling Edge; data is transmitted with the rising (XFE = 0) or falling edge (XFE = 1) of the device clock. Data stream manipulation according to these register entries only affects the system interface and only in the primary access configuration. The frame structure can be moved relative to the SP slope by up to 7-clock periods in half clock period steps. This register can hold non-zero values only for a CFR:CFS value of logical 0. Identical non-zero entries for RS2 – RS0 and XS2 – XS0 as well as identical RRE and XFE generate an output time-slot structure which is 1 time-slot late relative to the input time-slot structure. Identical 000 entries for RS2 – RS0 and XS2 – XS0 as well as RRE and XFE being logical 0 cause the input and output frames to coincide time. Figure 19 Clock Shifting Semiconductor Group 45 PEB 2245 5 Electrical Characteristics 5.1 Absolute Maximum Ratings Parameter Symbol Limit Values Unit Ambient temperature under bias TA 0 to 70 °C Storage temperature Tstg – 65 to 125 °C Voltage on any pin with respect to ground VS – 0.4 to VDD + 0.4 V Maximum voltage on any pin Vmax 7 V Note: Stresses above those listed here may cause permanent damage to the device. Exposure to absolute maximum rating conditions for extended periods may affect device reliability. 5.2 DC Characteristics TA = 0 °C to 70 °C; VDD = 5 V ± 5 %, VSS = 0 V. Parameter Symbol Limit Values min. max. Unit L-input voltage VIL – 0.4 0.8 H-input voltage VIH 2.0 VDD + 0.4 V L-output voltage VOL H-output voltage H-output voltage VOH VOH Power supply current Operational CLK = 4 MHz CLK = 8 MHz Input leakage current Output leakage current 0.45 2.4 VDD – 0.5 Test Condition V V IOL = 2 mA V V IOH = – 400 µA IOH = – 100 µA VDD = 5 V, ICC ICC 7 12 mA mA inputs at 0 V or VDD, no output loads ILI ILO 10 µA 0 V < VIN < VDD to 0 V 0 V < VOUT < VDD to 0 V Note: 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. 5.3 Capacitances TA = 25 °C, VDD = 5 V ± 5 %, VSS = 0 V. Parameter Symbol Limit Values min. Unit max. Input capacitance CIN 10 pF I/O capacitance CIO 20 pF Output capacitance COUT 15 pF Semiconductor Group 46 PEB 2245 5.4 AC Characteristics Ambient temperature under bias range, VDD = 5 V ± 5 %. Inputs are driven at 2.4 V for a logical 1 and at 0.4 V for a logical 0. Timing measurements are made at 2.0 V for a logical 1 and at 0.8 V for a logical 0. The AC testing input/output waveforms are shown below. 2.0 2.0 Device Under Test Test Points 0.8 0.8 C Load = 150 pf ITS00568 Figure 20 I/O Waveform for AC Tests Semiconductor Group 47 PEB 2245 Figure 21 Microprocessor Interface Timing Intel Bus Mode Semiconductor Group 48 PEB 2245 5.5 Microprocessor Interface Timing 5.5.1 Motorola Bus Mode Figure 22 µP Write Cycle Figure 23 Multiplexed Address Timing Figure 24 Non-multiplexed Address Timing Semiconductor Group 49 PEB 2245 Parameter Symbol Limit Values min. Unit max. ALE pulse width tAA 50 ns Address setup time to ALE tAL 20 ns Address hold time from ALE tLA 10 ns Address latch setup time to WR, RD tALS 0 ns Address setup time to WR, RD tAS 10 ns Address hold time from WR, RD tAH 20 ns ALE pulse delay tAD 15 ns DS delay after R/W setup tDSD 0 ns RD pulse width tRR 110 ns Data output delay from RD tRD 110 ns Data float from RD tDF 25 ns RD control interval tRI 70 ns WR pulse width tWW 60 ns Data setup time to WR + CS tDW 35 ns Data hold time from WR + CS tWD 10 ns WR control interval tWI 70 ns PCM-Interface Timing Parameter Symbol Limit Values min. Unit max. PCM-input setup tS 15 ns PCM-input hold tH 30 ns Output delay tD 45 ns Tristate delay tT 55 ns Semiconductor Group 50 PEB 2245 Clock and Synchronization Timing Parameter Symbol Limit Values min. Unit max. Clock period 8 MHz high tCP8 H 40 ns Clock period 8 MHz low tCP8 L 48 ns Clock period 8 MHz tCP8 120 ns Synchronization pulse setup 8 MHz tSS8 10 tCP8 – 20 ns Synchronization pulse delay 8 MHz tSH8 0 tCP8 – 20 ns Clock period 4 MHz high tCP4 H 90 ns Clock period 4 MHz low tCP4 L 90 ns Clock period 4 MHz tCP4 240 ns Synchronization pulse setup 4 MHz tSS4 10 tCP4 – 30 Synchronization pulse delay 4 MHz tSH4 30 tCP4 – 10 + tCP4H ns Data clock delay tDCD Synchronization pulse low tSPL Semiconductor Group 100 100 51 ns ns ns PEB 2245 1020 1021 1022 1023 0 1 2 3 CLK t CP 8H t CP 8L t SS 8 t SH 8 t CP 8 SP tS IN 2 Mbit/s tH TS 31, Bit 7 OUT 2 Mbit/s Time-Slot tS IN 4 Mbit/s TS 0 , Bit 0 tD 31, Bit 7 Time-Slot 0 , Bit 0 tH TS 63 , Bit 6 TS 63 , Bit 7 TS 0 , Bit 0 TS 0 , Bit 1 tD OUT 4 Mbit/s TS 63 , Bit 6 TS 63 , Bit 7 tS TS 127 Bit 4 IN 8 Mbit/s TS 127 Bit 5 TS 127 Bit 6 TS 0 , Bit 0 TS 0 , Bit 1 tH TS 127 Bit 7 TS 0 Bit 1 TS 0 Bit 0 TS 0 Bit 2 TS 0 Bit 3 tD TS 127 Bit 4 OUT 8 Mbit/s TS 127 Bit 5 TS 127 Bit 6 TS 0 Bit 0 TS127 Bit 7 TS 0 Bit 1 TS 0 Bit 2 Example with delayed output frame SP N = 255 OUT 2 Mbit/s TS 0 , Bit 0 TS 0 Bit 0 OUT 8 Mbit/s TS 0 Bit 1 TS 0 , Bit 1 TS 0 Bit 2 TS 0 Bit 3 Space switch application 0 1 TS 0 Bit 4 23 TS 0 Bit 5 TS 0 Bit 6 24 TS 0 Bit 7 280 281 SP IN 8 Mbit/s ~ ~~ ~ ~ ~~ ~ CLK tS N = 70 fixed TS 127 Bit 7 tH TS0 Bit 0 tD TS 127 , Bit 7 Figure 25 PCM-Line Timing in Standard Configuration with a 8-MHz Device Clock Semiconductor Group 52 TS 0 Bit 0 TS 0 Bit 0 ITT00592 PEB 2245 1020 1021 1022 1023 0 1 2 3 4 CLK SP t SS 8 Synchronous Interface IN tS IN 2 Mbit/s TS 31, Bit 7 System Interface IN IN 4 Mbit/s Synchronous Interface OUT System Interface OUT TS 0 , Bit 0 tH tS TS 31, Bit 7 IN 8 Mbit/s TS 63 , Bit 6 TS 63 , Bit 7 tS TS 127 Bit 4 TS 127 Bit 5 TS 0 , Bit 0 tH tS IN 8 Mbit/s t SH 8 tH TS 127 Bit 6 TS 0 , Bit 0 TS 0 , Bit 1 tH TS 127 Bit 0 TS127 Bit 7 TS 127 Bit 1 TS 127 Bit 2 TS127 Bit 3 tD OUT 2 Mbit/s Time-Slot 31, Bit 6 Time-Slot 0 , Bit 0 tD OUT 2 Mbit/s TS 31, Bit 7 TS 0 , Bit 0 tD OUT 4 Mbit/s TS 63 , Bit 6 TS 63 , Bit 7 TS 0 , Bit 0 TS 0 , Bit 1 tD OUT 8 Mbit/s TS 127 Bit 4 TS 127 Bit 5 TS 127 Bit 6 TS127 Bit 7 TS 127 Bit 0 TS 127 Bit 1 TS 127 Bit 2 System Interface TSC tT TSC 2 Mbit/s TS 31, Bit 7 TS 0 , Bit 0 tT TSC 4 Mbit/s TS 63 , Bit 6 TS 63 , Bit 7 TS 0 , Bit 0 TS 0 , Bit 1 tT TSC 8 Mbit/s TS 127 Bit 5 TS 127 Bit 6 TS 127 Bit 7 t DCD TS 127 Bit 0 TS 127 Bit 1 t DCD TS 127 Bit 2 ITT00593 Figure 26 PCM-Line Timing in Primary Access Configuration with a 8-MHz Device Clock and a CSR Entry (00010001) Semiconductor Group 53 PEB 2245 Figure 27 PCM-Line Timing in Standard Configuration with a 4-MHz Device Clock Semiconductor Group 54 PEB 2245 Figure 28 PCM-Line Timing in Primary Access Configuration with a 4-MHz Device Clock and a CSR Entry (00010001) Semiconductor Group 55 PEB 2245 Table 13 Busy Times Operation Max. Value Unit Indirect register access 900 ns Connection memory reset 250 µs Semiconductor Group 56 PEB 2245 6 Conference Applications of the MUSACTM Figure 29 Data Flow through the MUSACTM in Case of Conferencing Semiconductor Group 57 PEB 2245 The PCM samples of each input channel first pass through an input processing stage. In this stage, an input attenuation level (0, 3, 6 or 9 dB) and a noise suppression threshold can be programmed individually for each channel. Following the input processing the PCM data is expanded according to the A- or µ-law encoding rules and written to the Data Memory (DM). Additionally the PCM data of each input channel is added to the Conference Sum Memory (CSM). The DM location (1 out of 64) is specified by the conference control address (cca) and the CSM location (1 out of 21) is specified by the conference number when writing to the Conference Control Memory (CCM). The PCM data then passes through a substractor stage such that the resulting output channel for a given subscriber contains the contribution of all the other channels in the conference except its own. Finally the PCM data is forwarded to the output channel after PCM compression and an optional output attenuation of 3 dB. The common characteristics of all conferences (encoding law and byte format) as well as the conferencing function itself are selected in the configuration register CFR. The input channel and the individual conference parameters of a conference are programmed in the Conference Control Memory (CCM) which is a 20-bit (data) × 6-bit (address) memory. The 6-bit address, also called conference control address (cca), allows a maximum of 64 channels simultaneously involved in conference applications. The cca can be selected at random i.e. for each new participant of a conference any still not used cca can be taken. The 20-bit data field of the Conference Control Memory (CCM) specifies: – – – – – – the logical input line and time-slot number (9 bits), the input attenuation level (2 bits), the output attenuation level (1 bit), the noise suppression threshold (2 bits), the inversion function (1 bit) and the conference number (5 bits). The output channel and the conference mode are programmed in the Connection Memory (CM) which is an 11-bit (data) × 8-bit (address) memory. For each of the 256 output channels the 11-bit data field specifies the source and characteristics of the output line and time-slot selected by the CM address. Two cases can be distinguished: In the transparent switch mode 11 data bits specify: – the selection of the transparent switching mode (1 bit), – the output driver state (enabled/tristated) (1 bit) and – the logical input line and time-slot number (9 bits). In the conference switching mode 8 data bits (3 bits are not used) specify: – the selection of the conference switching mode (1 bit), – the output driver state, enabled or tristated (1 bit) and – the conference control address cca (6 bits). Semiconductor Group 58 PEB 2245 Procedure for Programming a Conference Configuration Register CFR ● The conference mode must be selected by setting CFR:FS = 1. ● The PCM encoding law must be selected, A-law (CFR:CUA0 = 1) or µ-law (CFR:CUA0 = 0). ● The PCM-byte format must be selected: CUA2 CUA1 PCM-Byte Format Application 0 0 1 1 0 1 0 1 No bits inverted Even bits inverted Odd bits inverted All bits inverted Required when using µ-law Required when using A-law Conference Control Memory For each input channel a CCM entry must be made to a dedicated conference control address (cca). This CCM entry defines the input line and time-slot number, and the input and output processing parameters. Writing to the CCM consists of two accesses: In the first access, the input channel is specified, in the second access, the input and output processing and the conference number are specified. Both accesses must always be performed in the sequence 1st access and then 2nd access. The Data Memory (DM) addressed by the cca buffers the input samples for output processing. The Conference Sum Memory (CSM), addressed by the conference number, contains the accumulated samples of all input channels having the same conference number. Connection Memory The output channels are programmed in the CM. For each output channel a CM entry must be made defining the output line and time-slot, enabling the output driver, and pointing to the same conference control address which has been used for the corresponding input channel. Each output channel contains the sum of the accumulated samples from the CSM minus the input samples retrieved from the DM of that same channel. Conference Overflow If one result of the final substraction exceeds the full-scale value, a saturation appears and an interrupt is generated (CST:COV = 1) to indicate this overflow condition. The conference number of the conference which caused the overflow is buffered in the Conference Status Register (CST:CN4 … 0). This information can then be used to program a larger attenuation level for the participants of that particular conference. Semiconductor Group 59 PEB 2245 Application Hints ● Connection and disconnection of individual participants from a conference: Subscribers can be connected and disconnected “on the fly” from a conference. A subscriber is disconnected from a conference by writing an invalid conference number (e.g. 15 h) to the CCM, by disabling the conference mode in the CM (D10 = 0), and by disabling the output driver in the CM (D9 = 1). ● Same subscriber participates in several different conferences at the same time: Input channels can also be used in several conferences at the same time. In this case several CCM entries with different conference control addresses and different conference numbers must point to the same input line and time-slot number. ● Same subscriber participates several times within the same conference: An input channel can also be used several times within the same conference. In this case several CCM entries with different conference control addresses must point to the same conference number and to the same input line and time-slot number. This function may serve for example to amplify the contribution of particular input channels within a conference. ● Tone insertion: A tone to be inserted into a conference can be programmed as an additional conference subscriber, but without assigning an output channel. For that purpose the CCM pointing to the desired conference number and to the input line and time-slot which contains the tone is written to, but no output channel is programmed in the CM. ● Switching of several input channels to one output channel: The conference function of the MUSAC can also be used to monitor the voice transmission of several telephone lines via a loudspeaker or for recording on tape. In this case several input channels must be “conference switched” to a single output channel. This output channel must contain the information of all input channels, i.e. a substraction of the input channel having the same conference control address (cca) as the output channel is not desired. For that purpose the following procedure can be adopted: For each input channel (e.g. in1 – in10) a CCM programming to individual conference control addresses is made (e.g. cca1 – cca10). All of these CCM entries must point to a unique conference number. One of the input channels, e.g. in1, is connected twice to that particular conference, i.e. a second CCM programming to another conference address (e.g. cca11) is made. For the desired output channel the CM is programmed to point to one of the two conference control addresses which have been used for input channel in1 (e.g. either cca1 or cca11). ● Inserting attenuation or noise suppression into a channel without conferencing: If an input channel shall be processed (e.g. attenuated) without being involved in a conference, the conference number 1FH can be used. Up to 64 input channels can then be individually processed by the internal DSP and switched to individual output channels. Each channel is assigned to a dedicated conference control address (0 – 63) but pointing to the unique conference number 31 (1FH). The processing parameters (level of attenuation, noise suppression threshold, etc.) are programmed in the CCM whereas the output channel is programmed in the CM. Semiconductor Group 60 PEB 2245 Meaning of Bits when Writing to the CCM and CM in Conference Mode A write access to indirect registers and to the CM and CCM is performed by writing a 3-byte sequence to the IAR register: 1st writing: 2nd writing: 3rd writing: IAR = control byte IAR = data byte IAR = address byte Before each indirect access, it should be verified that the STA:Z (incomplete instruction) and B bits (CM reset) are both set to logical 0. The Z bit is set to logical 1 after the first IAR access and reset to logical 0 at the latest 900 ns after the third access. An incomplete access sequence can be aborted by setting the MOD:RI bit to 1. Standard configuration and 2048-Mbit/s input and output modes are assumed for the encoding of time-slots and lines given below (for other configurations, refer to the MUSAC data sheet): CCM 1st access: CCM 2nd access: 0 1 1 0 0 0 INV ITS4 control byte ITS3 ITS2 ITS1 ITS0 IL3 IL2 IL1 IL0 data byte 0 0 CCA5 CCA4 CCA3 CCA2 CCA1 CCA0 address byte 1 0 1 0 0 0 NOI1 NOI0 control byte IAT1 IAT0 OAT CNR4 CNR3 CNR2 CNR1 CNR0 data byte 0 0 CCA5 CCA4 CCA3 CCA2 CCA1 CCA0 address byte Note that the CCM must always be written in the sequence: 1st access, 2nd access! CM access: 0 0 1 0 0 CONF=1 VAL 0 control byte 0 0 CCA5 CCA4 CCA3 CCA2 CCA1 CCA0 data byte OTS4 OTS3 OTS2 OTS1 OTS0 OL2 OL1 OL0 address byte INV PCM data is Inverted (1) or not (0) ITS4 … 0 Input Time-Slot number (0 – 31) IL3 … 0 Input Line number (0 – 15) CCA5 … 0 Conference Control Address (0 – 63) NOI1 … 0 Noise suppression threshold: 00 = no noise suppression 01 = 5th step, first segment 10 = 9th step, first segment 11 = 16th step, first segment IAT1 … 0 Input Attenuation: OAT Output Attenuation (0 = 0 dB, 1 = 3 dB) CNR4 … 0 Conference Number 0 … 20 valid conference number 31 no conference assigned 21 … 30 not used Semiconductor Group 00 = 0 dB 01 = 3 dB 10 = 6 dB 11 = 9 dB 61 PEB 2245 CONF Conference mode (1) or transparent mode (0) VAL Validity, output is enabled (0) or disabled (1) OTS4 … 0 Output Time-Slot number (0 – 31) OL2 … 0 Output Line number (0 – 7) Example of Programming a 4-Party Conference Configuration: standard configuration 2.048-Mbit/s input and output lines no inversion, no attenuation and no noise suppression conference number = 05 Subscriber 1: input channel 1: output channel 1: cca: line 0, time-slot 0 line 0, time-slot 0 02 Subscriber 2: input channel 2: output channel 2: cca: line 1, time-slot 8 line 1, time-slot 8 07 Subscriber 3: input channel 3: output channel 3: cca: line 4, time-slot 9 line 3, time-slot 5 01 Subscriber 4: input channel 4: output channel 4: cca: line 2, time-slot 7 line 0, time-slot 6 09 CCM Input Channel 1: CM Output Channel 1: W: IAR = 0110 0000 W: CCM 1st access W: IAR = 0000 0000 in I0, ts0 W: IAR = 0000 0010 cca = 02 W: IAR = 0010 0100 W: CM; conf. mode W: IAR = 0000 0010 cca = 02 W: IAR = 0000 0000 out I0, ts0 W: IAR = 1010 0000 W: CCM 2nd access W: IAR = 0000 0101 conf. nr. = 05 W: IAR = 0000 0010 cca = 02 CCM Input Channel 2: CM Output Channel 2: W: IAR = 0110 0000 W: CCM 1st access W: IAR = 1000 0001 in I1, ts8 W: IAR = 0000 0111 cca = 07 W: IAR = 0010 0100 W: CM; conf. mode W: IAR = 0000 0111 cca = 07 W: IAR = 0100 0001 out I1, ts8 W: IAR = 1010 0000 W: CCM 2nd access W: IAR = 0000 0101 conf. nr. = 05 W: IAR = 0000 0111 cca = 07 Semiconductor Group 62 PEB 2245 CCM Input Channel 3: CM Output Channel 3: W: IAR = 0110 0000 W: CCM 1st access W: IAR = 1001 0100 in I4, ts9 W: IAR = 0000 0001 cca = 01 W: IAR = 0010 0100 W: CM; conf. mode W: IAR = 0000 0001 cca = 01 W: IAR = 0010 1011 out I3, ts5 W: IAR = 1010 0000 W: CCM 2nd access W: IAR = 0000 0101 conf. nr. = 05 W: IAR = 0000 0001 cca = 01 CCM Input Channel 4: CM Output Channel 4: W: IAR = 0110 0000 W: CCM 1st access W: IAR = 0111 0010 in I2, ts7 W: IAR = 0000 1001 cca = 09 W: IAR = 0010 0100 W: CM; conf. mode W: IAR = 0000 1001 cca = 09 W: IAR = 0011 0000 out I0, ts6 W: IAR = 1010 0000 W: CCM 2nd access W: IAR = 0000 0101 conf. nr. = 05 W: IAR = 0000 1001 cca = 09 Disconnecting Subscriber 3 from the Conference: CCM Input Channel 3: CM Output Channel 3: W: IAR = 0110 00xx W: CCM 1st access W: IAR = xxxx xxxx don’t care W: IAR = 0000 0001 cca = 01 W: IAR = 0010 001x W: IAR = 1010 00xx W: CCM 2nd access W: IAR = xxx1 0101 invalid conf. nr. = 21 (15h) W: IAR = 0000 0001 cca = 01 Semiconductor Group 63 W: CM; transp. mode, output disabled W: IAR = xxxx xxxx don’t care W: IAR = 0010 1011 out I3, ts5 PEB 2245 Reading Back the CCM and the CM Contents of Subscriber 2: CCM Input Channel 2: CM Output Channel 2: W: IAR = 0100 0000 R: CCM 1st access (K2 – K0 = 010) W: IAR = 0000 0000 data don’t care e.g. 0 W: IAR = 0000 0000 R: CM command (K2 – K0 = 000) W: IAR = 0000 0000 data don’t care e.g. 0 W: IAR = 0100 0001 address is out I1, ts8 R: IAR = 0010 0100 only D10 – D8 bit positions are valid R: IAR = 0000 0111 read back of D7 – D0 (cca = 07) R: IAR = 0100 0001 read back of address (out I1, ts8) W: IAR = 0000 0111 address is cca = 07 W: IAR = 0100 0000 R: CCM 1st access (K2 – K0 = 010) W: IAR = 0000 0000 data don’t care e.g. 0 W: IAR = 0000 0111 address is cca = 07 R: IAR = 0100 0100 only D9 – D8 are valid R: IAR = 1000 0001 read back of D7 – D0 (in I1, ts8) R: IAR = 0000 0111 read back of address (cca = 07) W: IAR = 1000 0000 R: CCM 2nd access (K2 – K0 = 100) W: IAR = 0000 0000 data don’t care e.g. 0 W: IAR = 0000 0111 address is cca = 07 W: IAR = 1000 0000 R: CCM 2nd access (K2 – K0 = 100) W: IAR = 0000 0000 data don’t care e.g. 0 W: IAR = 0000 0111 address is cca = 07 R: IAR = 1010 0000 only D9 – D8 are valid R: IAR = 0000 0101 read back of D7 – D0 (conf. nr. = 05) R: IAR = 0000 0111 read back of address (cca = 07) Semiconductor Group 64 PEB 2245 7 Package Outlines GPL05102 P-LCC-44 (SMD) (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 65 Dimensions in mm PEB 2245 8 Appendix 8.1 Initialization for Conferencing in a PBX 1) 2) See table 7. See table 8. Figure 30 Semiconductor Group 66 PEB 2245 8.2 Programming a Conference in a PBX Figure 31 Table 14 Procedure for each Conference Bits in IAR Values for Subscribers TE ‘A’ 1. Make a list of the conference subscribers 2. Determine the input time slot Determine the logical input port ITS6 … 0 IL3 … 0 3. Determine the output time slot Determine the logical output port OTS6 … 0 OL2 … 0 4. Find an unused conference number CNR4 … 0 5. Assign an unused conference control address to every conference input CCA5 … 0 6. Fix the input attenuation IAT1 … 0 7. Fix the output attenuation OAT 8. Fix the noise suppression threshold NOI1 … 0 9. Insert the determined values in the corresponding bit positions of the ‘MUSAC-A Work Sheet’ (see chapter 8.3 and 8.4). Write the 9 bytes to the CM and CCM using the three byte IAR access. Semiconductor Group 67 TE ‘B’ TE ‘C’ PEB 2245 8.3 Programming Procedure for Switching TS’s – Select a column for input and output rate – Fill in the values of the bits – Write the 3 bytes (from top to bottom) to register IAR 2 MBit/s Output Rate 4 MBit/s Output Rate 8 MBit/s Output Rate 3 byte IAR access to CM: IAR IAR 0 1 0 0 0 0 ITS3…0 IAR VAL ITS4 IL3…0 OTS4…0 OL2…0 IAR IAR IAR 0 0 1 0 ITS4…0 OTS5…0 ITS6 … 0 IL0 … 3 Input time slot number Logical input line number VAL Validity: OTS6 … 0 Output time slot OL2 … 0 Logical output line number Semiconductor Group 0 output enabled (= 0) output disabled (= 1) 68 0 VAL ITS5 IL2…0 OL1…0 IAR IAR IAR 0 0 1 0 ITS5…0 OTS6…0 0 0 VAL ITS6 IL1…0 OL0 PEB 2245 8.4 Programming Procedure for a PBX Conference – Select a column for input and output rate – Fill in the values of the bits by aid of chapter 8.1 and 8.2 – Write the 9 bytes (from top to bottom) to register IAR 2 MBit/s Input Rate 4 MBit/s Input Rate 8 MBit/s Input Rate First 3 byte IAR access to CCM: IAR IAR 0 1 1 0 0 ITS3 … 0 0 INV ITS4 IL3 … 0 IAR 0 1 1 0 0 0 ITS4 … 0 IAR IAR 0 0 IAR 1 0 INV ITS5 IL2 … 0 IAR IAR 0 1 1 0 0 ITS5 … 0 0 INV ITS6 IL1 … 0 CCA5 … 0 second 3 byte IAR access to CCM: IAR IAR 2 MBit/s Output Rate 1 0 IAT1 … 0 OAT 0 0 0 NOI1 … 0 CNR4 … 0 CCA5 … 0 0 4 MBit/s Output Rate 8 MBit/s Output Rate 3 byte IAR access to CM: IAR OTS4 … 0 Semiconductor Group OL2 … 0 IAR 0 0 IAR 0 0 IAR 1 0 0 VAL 1 0 CCA5 … 0 OTS5 … 0 69 OL1 … 0 IAR OTS6 … 0 OL0 PEB 2245 INV PCM data inverted (= 1) or not (= 0) ITS6 … 0 Input time slot number IL0 … 3 Logical input line number CCA5 … 0 Conference control address NOI1 … 0 Noise suppression threshold: 00 = no noise suppression 01 = 5th step, first segment 10 = 9th step, first segment 11 = 16th step, first segment IAT1 … 0 Input attenuation: 00 = 0 dB 01 = 3 dB 10 = 6 dB 11 = 9 dB OAT Output attenuation: 0 = 0 dB, 1 = 3 dB CNR4 … 0 Conference number VAL Validity: output enabled (= 0) output disabled (= 1) OTS6 … 0 Output time slot OL2 … 0 Logical output line number Semiconductor Group 70