Preliminary Technical Information

AD6426

This Information applies to a product under development. Its characteristics and specifications are subject to change without notice. Analog Devices assumes no

obligation regarding future manufacture unless otherwise agreed to in writing. No responsibility is assumed by Analog Devices for its use; nor for any

infringements of patents or other rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent

rights of Analog Devices.

Revision Preliminary 2.3 (June 9, ´98)

- 7 -

Confidential Information

OVERVIEW

The GSM air interface has been formulated to provide high

quality digital mobile communication. As well as supporting

the traffic channels (speech and/or data), the air interface

specifies a number of signaling channels that are used for call

set up and communications between the network infrastructure

and the mobile. These signaling channels provide the mobile

specific features such as handover, as well as a number of

other intelligent features.

The GSM system closely follows the OSI 7-layer model for

communications. Specifically, GSM defines Layers 1, 2 and 3

of the protocols. The lowest level being Layer 1, or the

Physical Layer. It is this part of the network processing for

which the EGSMP is responsible, performing some of the

Layer 1 functions in dedicated hardware for minimum power

consumption and some in software for increased flexibility.

Layer 1 covers those signal processing functions required to

format the speech/data for transmission on the physical

medium. Data must be structured to allow for identification,

recovery and error correction so that the information can be

supplied error free to the layer 2 sub-systems and to the traffic

sources. In addition, the physical layer processing includes the

timing of both transmit and receive data, the encryption of

data for security purposes and the control of the Radio sub-

system to provide timing and to optimize the radio frequency

characteristics. An object code license to Layer 1 software is

supplied with the AD20msp425 chipset.

FUNCTIONAL PARTITIONING

This datasheet gives only an overview about the functionality

of the EGSMP. The EGSMP consists of three main elements;

the Channel Codec and the Control Processor Sub-System

including several interfaces and the DSP as shown in

Figure

coding and decoding of traffic and control information. The

Processor Sub-system supports the software functions of the

protocol stack and interfaces with the bus peripheral sub-

systems of the terminal. The DSP performs the channel

equalization and speech transcoding.

Channel Codec Sub-System

The Channel Codec processes data from two principal sources;

traffic and signaling. The former is normally continuous and

the latter determined on demand. Traffic comes in two forms;

speech and user data. The various traffic sources and the

signaling sources are all processed differently at the physical

layer. Speech traffic data is supplied by the speech transcoder

and the remaining data types are sourced from the Control

Processor and interfaced via a dedicated data interface. The

Channel Codec subsystem functional block diagram is shown

in Figure 3.

DSP

INTERFACE

REGISTERS

DEINTERLEAVE

INTERLEAVE

ENCODE

DECODE

ENCRYPT

DECRYPT

VBC

INTERFACE

TEST

INTERFACE

H8

INTERFACE

RADIO / SYNTHESIZER

TIMING AND CONTROL

Figure 3. Channel Codec Subsystem

The transmit and receive functions of the Channel Codec are

timed by an internal timebase that maintains accurate timing

of all sub-systems. This timebase is aligned with the on-air

receive signal and all system control signals, both internal and

external, are derived from it.

The physical layer processing can be divided into 4 phases,

two each for up- and downlink. The data in the transmit path

undergoes an ENCODE phase and then a TRANSMIT phase.

Similarly, data in the downlink path is termed the receive data

and it undergoes a RECEIVE phase followed by a DECODE

phase. The buffer between the ENCODE and TRANSMIT

functions is the INTERLEAVE module that holds the data and

permits the building of the transmit burst structure. Similarly

the DEINTERLEAVE module forms the buffer between the

RECEIVE and the DECODE processes.

Each of these four phases is controlled explicitly by the

Control Processor via control registers that define the mode of

operation of each sub-module and the data source they should

process. Typically these control values are updated every

TDMA frame in response to interrupts from the internal

timebase.

The ENCODE process involves the incorporation of error

protection codes. All data is sourced in packets and two forms

of error coding applied; block coding (parity or Fire code) and

convolution coding. The resultant data block is then written to

the INTERLEAVE module where it is buffered in a RAM.

Data is read from the interleave buffer memories contiguously

but written in non-contiguous manner, thereby implementing

the interleaving function. The TRANSMIT process uses a

different time structure now associated with the on-air TDMA

structure. The data is read from the INTERLEAVE module

and formatted into bursts with the requisite timing. This

involves adding fixed patterns such as the tail bits and training

sequence code. The resultant burst is written to the external

Baseband Converter where the modulation is performed and

the output timed to the system timebase before transmission.