Computer Simulation and DSP Implementation of Data Mappers of V.90 Digital Modem in Theaid of IT

Similar documents
INTERNATIONAL TELECOMMUNICATION UNION. SERIES V: DATA COMMUNICATION OVER THE TELEPHONE NETWORK Simultaneous transmission of data and other signals

INTERNATIONAL TELECOMMUNICATION UNION. SERIES V: DATA COMMUNICATION OVER THE TELEPHONE NETWORK Simultaneous transmission of data and other signals

Visa Smart Debit/Credit Certificate Authority Public Keys

Secret Key Systems (block encoding) Encrypting a small block of text (say 128 bits) General considerations for cipher design:

C Mono Camera Module with UART Interface. User Manual

SEN366 Computer Networks

Data Encoding g(p (part 2)

ETSI TS V ( )

10 Speech and Audio Signals

CSci 127: Introduction to Computer Science

Waveform Encoding - PCM. BY: Dr.AHMED ALKHAYYAT. Chapter Two

Overview. Chapter 4. Design Factors. Electromagnetic Spectrum

6.450: Principles of Digital Communication 1

Signal Encoding Techniques

Mansour Keramat. * No part may be reproduced without permission from the author. 1- Application of Data Converters. Contents

CS420/520 Axel Krings Page 1 Sequence 8

Function Block DIGITAL PLL. Within +/- 5ppm / 10 years (Internal TCXO Stability) 1 External Reference Frequency Range: 10MHz +/- 100Hz

Data Communications and Networking (Module 2)

Multiplexing. Chapter 8. Frequency Division Multiplexing Diagram. Frequency Division Multiplexing. Multiplexing

xdsl Modulation Techniques

The Physical Layer Outline

Adoption of this document as basis for broadband wireless access PHY

Voice Transmission --Basic Concepts--

Next: Broadcast Systems

William Stallings Data and Computer Communications. Chapter 8 Multiplexing. Multiplexing

Class 4 ((Communication and Computer Networks))

Mobile Communications Chapter 6: Broadcast Systems

EEE 309 Communication Theory

Recommendation ITU-R BT.1577 (06/2002)

Chapter-1: Introduction

ET4254 Communications and Networking 1

Digital Communication Systems. Asymmetric Digital Subscriber Line (ADSL) Gavin Cameron

Department of Electronics and Communication Engineering 1

Data and Computer Communications. Tenth Edition by William Stallings

Signal Encoding Techniques

Digital Lighting Systems, Inc. PD804-DMX. Eight Channel DMX Pack. (includes information for PD804-DMX-S) USER'S MANUAL. PD804-DMX-UM Rev.

ON SYMBOL TIMING RECOVERY IN ALL-DIGITAL RECEIVERS

UNIT-1. Basic signal processing operations in digital communication

MTI 7603 Pseudo-Ternary Codes

TELECOMMUNICATION SYSTEMS

a. Find the minimum number of samples per second needed to recover the signal without loosing information.

ENGR 4323/5323 Digital and Analog Communication

Audit Attestation Microsec ETSI Assessment 2017 No. AA

Fundamentals of Digital Communication

8WD4 Signaling Columns

Audit Attestation for SwissSign AG. This is to confirm that TUV AUSTRIA CERT has successfully audited the CAs of SwissSign without critical findings.

In this lecture. System Model Power Penalty Analog transmission Digital transmission

Chapter 12: Digital Modulation and Modems

) #(2/./53 $!4! 42!.3-)33)/.!4! $!4! 3)'.!,,).' 2!4% ()'(%2 4(!. KBITS 53).' K(Z '2/50 "!.$ #)2#5)43

P. 241 Figure 8.1 Multiplexing

EEE 309 Communication Theory

9/24/08. Broadcast Systems. Unidirectional distribution systems. Unidirectional distribution. Unidirectional distribution systems DAB Architecture

745 Transformer Protection System Communications Guide

Lecture 9b Convolutional Coding/Decoding and Trellis Code modulation

TCM-coded OFDM assisted by ANN in Wireless Channels

DEGEN DE1103 FM / MW / SW RECEIVER FM / AM / SSB / CW MODES OPERATING MANUAL

Data Communications and Networks

SIGNAL CLASSIFICATION BY DISCRETE FOURIER TRANSFORM. Pauli Lallo ABSTRACT

)454 ' ).4%27/2+).' "%47%%..%47/2+3 "!3%$ /. $)&&%2%.4 $)')4!, ()%2!2#()%3!.$ 30%%#( %.#/$).',!73 $)')4!,.%47/2+3. )454 Recommendation '

Time division multiplexing The block diagram for TDM is illustrated as shown in the figure

CARLETON UNIVERSITY Department of Systems and Computer Engineering

EXPERIMENT WISE VIVA QUESTIONS

Data and Computer Communications Chapter 8 Multiplexing

CSCD 433 Network Programming Fall Lecture 5 Physical Layer Continued

PULSE CODE MODULATION (PCM)

Chapter 2: Fundamentals of Data and Signals

Implementation of Different Interleaving Techniques for Performance Evaluation of CDMA System

EE 304 TELECOMMUNICATIONs ESSENTIALS HOMEWORK QUESTIONS AND ANSWERS

Lecture 3 Concepts for the Data Communications and Computer Interconnection

Wireless Communication Fading Modulation

CHAPTER 3 Syllabus (2006 scheme syllabus) Differential pulse code modulation DPCM transmitter

EC 2301 Digital communication Question bank

CSE 123: Computer Networks Alex C. Snoeren. Project 1 out Today, due 10/26!

ECE230X Lectures 10-11

EFFECTIVE CHANNEL CODING OF SERIALLY CONCATENATED ENCODERS AND CPM OVER AWGN AND RICIAN CHANNELS

Pulse Code Modulation

Chapter 4 Digital Transmission 4.1

16QAM Symbol Timing Recovery in the Upstream Transmission of DOCSIS Standard

Physical Layer. Networks: Physical Layer 1

Predicting Total Harmonic Distortion (THD) in ADSL Transformers using Behavioural Modeling

2. By convention, the minimum and maximum values of analog data and signals are presented as voltages.

Supplier s declaration of conformity

Data Communications and Networks

BER Performance Comparison between QPSK and 4-QA Modulation Schemes

Access to Data & Computer Networks Physical Level

Chapter 2 Overview - 1 -

Chapter 4 Digital Transmission 4.1

Performance Analysis of OFDM System with QPSK for Wireless Communication

DIGITAL COMMUNICATION

Lecture 2: Links and Signaling"

Stream Information. A real-time voice signal must be digitized & transmitted as it is produced Analog signal level varies continuously in time

Computer Networks

EITF25 Internet Techniques and Applications L2: Physical layer. Stefan Höst

ECE 271 INTRODUCTION TO TELECOMMUNICATION NETWORKS HOMEWORK QUESTIONS ECE 271 HOMEWORK-1

Performance Analysis of n Wireless LAN Physical Layer

Pulse Code Modulation (PCM)

Part VI: Requirements for ISDN Terminal Equipment

techniques are means of reducing the bandwidth needed to represent the human voice. In mobile

Chapter 3 Digital Transmission Fundamentals

Part VI: Requirements for Integrated Services Digital Network Terminal Equipment

Transcription:

Asian Journal of Information Technology 4 (6): 600-606, 2005 Grace Publications, 2005 Computer Simulation and DSP Implementation of Data Mappers of V.90 Digital Modem in Theaid of IT Jasvir Singh and Davinderpal Sharma Department of Electronics Technology, Guru Nanak Dev University, Amritsar, Punjab, India Abstract: Voice band modem operating at the rate of 56 Kbps is gaining popularity these days. This study presents computer-aided simulation and implementation of Data Mappers for V.90 digital modem transmitter. The Present work deals with the development of an efficient algorithm for Data Mappers used in V.90 digital modem transmitter, Corresponding assembly program development (source code generation), its simulation on simulator of Code Composer Studio and finally implementation of Data Mappers on Digital Signal Processor (DSP) have been discussed. Implementation parameter like total program execution time, data memory and program memory used for the present implementation has been presented. Key words: Information Technology (IT), Digital signal processor, digital modem, data mappers INTRODUCTION Modern telecommunications are undergoing dramatic changes. The merging of communication and computing devices and wide use of Internet offer users to enjoy teleconferencing, World Wide Web surfing, Internet phone, Movies and e-commerce. This has become possible due to the device so called Modem, derived from combinations of its two major function modulation and demodulation. Over the past few years, it has been repeatedly predicted that network access via telephone lines would be replaced by new services based on emerging technologies [1]. Despite all these predictions, voiceband modems [2,3] are still used by the majority of home computer users and small business owners for data communication and network access. There is no doubt that the demand for more reliable and efficient information access will bring changes to this situation. Several competing technologies are available to provide the solution to the so-called last mile problem, i.e. how to connect customers premises to the broadband network in a more flexible and affordable way. Services such as Integrated System Digital Network (ISDN) and Asymmetric Digital Subscriber Line (ADSL) are two examples of digital data transmission over the twisted-pair subscriber lines. Compared to traditional analog telephone lines, these technologies require investment by the telephone companies as to install special equipment at the central offices. Depending on the distance between user premises and the serving central office, some users may not be able to use such services. Other technologies for network access such as coaxial cable, wireless access, direct satellite access and fiber optic lines are also available [4]. Technologies developed for each of these media have shown merits for particular applications. However, there seems to be no universal technology for all locations and all applications. Among all access technologies, the POT Service provides the widest access coverage in the world. In accordance with a study made by Geogia Tech, in 1998 approximately 70% of Internet users were connected to the network with analog voiceband modems and according to the survey conducted by a firm (Jupiter Communication) more than 50 million people in the US alone were using telephone dial up technology to access the Internet in 2001 and the user strength is growing up continuously [5]. The Gartner Group estimates that about 55% of the user will rely on voiceband services even till 2004 [6]. Moreover voiceband modem has many advantages over others. Few of them are given below: Inexpensive Easy to install More reliable Easy functioning Widely available. Seeing the huge consumer market of voiceband modem for Internet access, present study was carried out on data transmission over analog telephone lines. V.90/V.92 is the current modem standard over the PSTN telephone lines, which uses entirely different technology. Block diagram of V.90 modem communication system is shown in Fig. 1. Traditional analog modem like V.34 assumes both the ends of the modem session to have an analog connection to PSTN whereas V.90 standard assumes one end of the modem section to be purely digital to take the advantage of high-speed connection [7]. Internet Service Providers Corresponding Author: Jasvir Singh, Department of Electronics Technology, Guru Nanak Dev University, Amritsar, Punjab, India E-mail: Jasvirsingh00@yahoo.com 600

Fig. 1: Block diagram of V.90 system (ISPs) are already using digital connection at their end. There is only one analog portion on the downstream transmission path (from ISP to DTE) and the upstream data conforms to the V.34 standard. TCM is used in upstream direction whereas in downstream Pulse Coded Modulation (PCM) as specified by ITU in G.711 recommendation [8] is used in V.90 modem, also known as PCM modem. V.90 digital modem transmitter: Transmitter of V.90 as well as V.92 digital modem is shown in the Fig. 2. First unit is scrambler whose purpose is to facilitate effective transmission of the data over the telephone channel and to improve the convergence of the adaptive equalization and echo cancellation in the receiver. It helps the receiver to recover the timing information from the received data to facilitate synchronous operation. The downstream encoder in Draft Recommendation V.90 uses multiple modulus conversion as its mapping scheme and convolutional spectral shaping as its spectral shaping scheme. The block diagram in Fig. 3 shows an overview of the downstream encoder and represents one data frame. Data frames in the digital modem have a six-symbol structure (since the robbed-bit signaling pattern repeats every six symbols). Each symbol position within the data frame is called a data frame interval and is indicated by a time index, i = 0,..., 5. During startup, the following encoder parameters are established: C i equals the positive constellation points for data frame interval i. M i is the number of code points in each constellation C i. K is the number of modulus encoder input data bits per data frame. S r is the number of PCM code sign bits per data frame used as redundancy for spectral shaping. S is the number of differential encoder input data bits per data frame, where S + S r = 6. The positive constellations (C i ) to be used in each data frame interval are specified by the analog modem during training procedures. The signaling rate is 601 Fig. 2: Transmitter of V.90 digital modem (server side) D S = [(K+S) * 8000]/6 (1) determined by the selection of the parameters K and S during the startup phase using formula given by eqn. 1. Description of each of the components or functional blocks as presented in Fig. 3 is given below [7] : Bit Parser partitions the block of binary data for one mapping frame into different groups of bits for processing by subsequent stages of the transmitter. It takes bits from scrambled data stream and parses them into two groups, which are fed to two different parts of encoder i.e. to differential encoder and modulus encoder. It takes D (equal to S + K) scrambled input data bits (d 0 : d D-1 ) and parses them into K modulus encoder input bits (b 0 : b K-1 ) and S differential encoder input bits ( s 0 : s S- 1 ). The modulus encoder takes K bits from the bit parser and maps them into six integers K i, where i=1,2,,6. Each K i is an integer between 0 and M i, where the M i s are called the mapping moduli and represent the number of elements in each of the PCM code sets defined for data frame interval 0 to data frame interval 5. In order to be able to represent the information in the K bits taken from the parser with these six integers, the values of M i and K must satisfy the following inequality: Each frame interval has an independent mapper associated with it. Each one of them also has a tabulation of M i PCM codes corresponding to the positive elements of the constellation to be used by it and denoted by C i. The specific PCM codes that assemble each of the constellations are selected by the analog modem during the startup phase of the communication. It is required that the members of C i should be labeled in descending order so the label 0 corresponds to the largest PCM code in the constellation and the label M i correspond to the smallest. The output of each mapper (U i ) is generated by selecting the constellation point in C i corresponding to K i. The S differential encoder input bits (s 0 : s S-1 ) are parsed into j = S r spectral shaping frames of length 6/S r, where the parser output j (n) represents the n th bit of the j th spectral shaping frame in a data frame. The six p j (n) bits are then differentially encoded to produce six input sign bits, t j (n), to the spectral shaper.

Fig. 3: V.90 digital modem encoder The spectral shaping is intended to change the shape of the spectrum of the transmitted signal to make it better suited to the channel used. Spectral shaping affects only the signs of the transmitted PCM symbols. The spectral 5 2 K = M i (2) i=0 shaper modifies input sign bits t j (n) to corresponding PCM code-sign bits ($ i ) so as to minimize the spectral shaping metric without violating the constraint specified in V.90 recommendation [7,9]. The six sign bits generated by the spectral shaper ($ i ) are attached to the six unsigned mapper outputs (U i ) to form the six output symbols (PCM i ), which are then multiplexed to form the stream of PCM octets to be transmitted. This completes the encoding process. 8-bit PCM codes generated by the transmitter arrive at the central telephone office through the internal digital telephone network and are applied to the digital to analog converter in the Codec at the rate of 8000 samples per second. The Codec converts each code to one of 256 voltage levels and passes the resulting staircase waveforms through a lowpass filter with a 4kHz cut-off frequency [10]. The linear to µ/a-law Converter, who expands the 8 encoded PCM bits to 14 bits in accordance with the ITU recommendation G.711 [8]. The procedure of expanding 8-bit input to 14-bit data at transmitter and the compressing the 14-bit data to 8-bit at other end is called Companding. The device, which accomplishes this task, is called CODEC and is generally situated at 602 the central office. A-law is used by European countries whereas in U.S.A. µ-law popular. Low pass filter in the modem design is generally used to avoid the aliasing problem caused by ADC in the communication path. To avoid the aliasing problem it should be ensured that the ADC never sees any signals that are to high in frequency. This is also known as antialiasing filter. As discussed above this filter has cut-off frequency equal to the bandwidth of the channel used i.e. 4kHz. The output of filter is connected to the twisted pair (telephone line) through the hybrid circuit installed at the local telephone office of a client. Data mapping: V.90 modem system uses a data transmission technique that performs Multiple Modulus Conversion (MMC) on a number of bits provided by the scrambler or other processing elements like parser. MMC is a technique for expressing an integer number as a sum of quotients containing multiple bases or moduli. MMC techniques, which are currently being used in V.90 modem systems, are inefficient from implementation point of view due to involvement of large number of computations like long word division. These techniques also require large amount of calculations and program code development time and apart from these, such MMC process are difficult to implement in modems where high data transmission rates are required as in the present case. Modulus encoder is a device that perform this MMC process having first means for obtaining a number of moduli, second means for obtaining a binary input having multiple bits and a processor suitably configured to

represent the binary input as a plurality of short words and perform short word division procedure in connection with the short words to thereby determine a plurality of index values associated with the moduli. The efficiency of the MMC process generally depends upon the bit capacity of the DSP used in the system. Division routines in fixed point DSP as used in present case are generally inefficient so present study introduces an algorithm that enables MMC processing of a large number of input bits (greater than bit capacity of DSP) using DSP that have a limited bit processing capacity. Data Mapper takes K i bits from modulus encoder and forms U i codes by choosing the constellation point in C i labelled by K i as shown in Fig. 3. There are six independent data mappers associated with the six data frame intervals. Each mapper uses a tabulation of M i PCM code that make up the positive constellation points of data frame interval 'i' denoted C i. The PCM codes to be used in each data frame interval is specified by the analog modem during training procedure [7]. Block diagram of Data Mapper is given in Fig.4. C i corresponds to the Universal set of PCM code words as specified by the ITU Recommendation V.90 in Table 1/V.90 [7]. This table contains both A-law and??law PCM code along with their linear codes. As A-law is being followed in India so C i values have been chosen from the A-law Universal Set of PCM codewords. Asian J. Inform. Technol., 4(6): 600-606, 2005 Algorithm for implementation of data mappers: Figure 5 shows the flow diagram of the algorithm being used in the implementation of Data Mappers functionality. The PCM code set members C i have been labelled in descending order so that label '0' corresponds to the largest PCM code in C i and label M i 1 corresponds to the smallest PCM code in C i as required by the ITU Recommendation V.90. In the very beginning 128 universal PCM code words corresponding to A-law (as given in Appendix (A)), which forms the positive side of constellation, can be stored at appropriate program memory address (pma), Fig. 4: Block diagram of data mapper of V.90 digital modem Fig.5: Algorithm to implement data mappers for transmitter of V.90 digital modem which in the next step can be shifted to the data memory to achieve Data Mapping function. PCM code words can also be directly loaded in to the appropriate 'dma'. In the next step initialization of data mapping frame can be done by setting mapper index to zero which means that data mapping frame '0' is being processed or data mapper '0' is being activated. After setting the data mapper for a particular data frame, output from modulus encoder K i corresponding to the same data frame can be fed to data mapper which maps K i = C i e.g. if K 0 =10 (output of modulus encoder during 0 th data frame interval) then data mapper 0 will select PCM code word corresponding to U-code (C i )=10. In the next step selected PCM code word C i can be stored at appropriate dma U i or given to the next processing unit of the transmitter. Query task is performed to activate the next data mapper corresponding to next data frame and the data mapping process described above is repeated till all the six data mapper gets activated and their corresponding outputs get stored at 'dma' U i. 603

Appendix (A) The Universal set of PCM Code Words U-code? -law? -law A-law A-law U-code? -law? -law A-law A-law PCM linear PCM linear PCM linear PCM linear 0 FF 0 D5 8 64 BF 1980 95 2112 1 FE 8 D4 24 65 BE 2108 94 2240 2 FD 16 D7 40 66 BD 2236 97 2368 3 FC 24 D6 56 67 BC 2364 96 2496 4 FB 32 D1 72 68 BB 2492 91 2624 5 FA 40 D0 88 69 BA 2620 90 2752 6 F9 48 D3 104 70 B9 2748 93 2880 7 F8 56 D2 120 71 B8 2876 92 3008 8 F7 64 DD 136 72 B7 3004 9D 3136 9 F6 72 DC 152 73 B6 3132 9C 3264 10 F5 80 DF 168 74 B5 3260 9F 3392 11 F4 88 DE 184 75 B4 3388 9E 3520 12 F3 96 D9 200 76 B3 3516 99 3648 13 F2 104 D8 216 77 B2 3644 98 3776 14 F1 112 DB 232 78 B1 3772 9B 3904 15 F0 120 DA 248 79 B0 3900 9A 4032 16 EF 132 C5 264 80 AF 4092 85 4224 17 EE 148 C4 280 81 AE 4348 84 4480 18 ED 164 C7 296 82 AD 4604 87 4736 19 EC 180 C6 312 83 AC 4860 86 4992 20 EB 196 C1 328 84 AB 5116 81 5248 21 EA 212 C0 344 85 AA 5372 80 5504 22 E9 228 C3 360 86 A9 5628 83 5760 23 E8 244 C2 376 87 A8 5884 82 6016 24 E7 260 CD 392 88 A7 6140 8D 6272 25 E6 276 CC 408 89 A6 6396 8C 6528 26 E5 292 CF 424 90 A5 6652 8F 6784 27 E4 308 CE 440 91 A4 6908 8E 7040 28 E3 324 C9 456 92 A3 7164 89 7296 29 E2 340 C8 472 93 A2 7420 88 7552 30 E1 356 CB 488 94 A1 7676 8B 7808 31 E0 372 CA 504 95 A0 7932 8A 8064 32 DF 396 F5 528 96 9F 8316 B5 8448 33 DE 428 F4 560 97 9E 8828 B4 8960 34 DD 460 F7 592 98 9D 9340 B7 9472 35 DC 492 F6 624 99 9C 9852 B6 9984 36 DB 524 F1 656 100 9B 10364 B1 10496 37 DA 556 F0 688 101 9A 10876 B0 11008 38 D9 588 F3 720 102 99 11388 B3 11520 39 D8 620 F2 752 103 98 11900 B2 12032 40 D7 652 FD 784 104 97 12412 BD 12544 41 D6 684 FC 816 105 96 12924 BC 13056 42 D5 716 FF 848 106 95 13436 BF 13568 43 D4 748 FE 880 107 94 13948 BE 14080 44 D3 780 F9 912 108 93 14460 B9 14592 45 D2 812 F8 944 109 92 14972 B8 15104 46 D1 844 FB 976 110 91 15484 BB 15616 47 D0 876 FA 1008 111 90 15996 BA 16128 48 CF 924 E5 1056 112 8F 16764 A5 16896 49 CE 988 E4 1120 113 8E 17788 A4 17920 50 CD 1052 E7 1184 114 8D 18812 A7 18944 51 CC 1116 E6 1248 115 8C 19836 A6 19968 52 CB 1180 E1 1312 116 8B 20860 A1 20992 53 CA 1244 E0 1376 117 8A 21884 A0 22016 54 C9 1308 E3 1440 118 89 22908 A3 23040 55 C8 1372 E2 1504 119 88 23932 A2 24064 56 C7 1436 ED 1568 120 87 24956 AD 25088 57 C6 1500 EC 1632 121 86 25980 AC 26112 58 C5 1564 EF 1696 122 85 27004 AF 27136 59 C4 1628 EE 1760 123 84 28028 AE 28160 60 C3 1692 E9 1824 124 83 29052 A9 29184 61 C2 1756 E8 1888 125 82 30076 A8 30208 62 C1 1820 EB 1952 126 81 31100 AB 31232 63 C0 1884 EA 2016 127 80 32124 AA 32256 604

Simulation of data mappers: Program corresponding to the algorithm described above has been developed in assembly language of TMS320C50PQ57 DSP to perform data mapping function in transmitter of V.90 igital modem. The source codes corresponding to the data mapper function have been developed. These source codes after converting in to appropriate format have been loaded in to the simulator of code composer studio where assembly program for data mappers has been debugged. The status of the simulator before execution of the program has been shown in Fig. 6. As an example it was assumed that the output of modulus encoder (K i ) are as follows: K 0 = 10 ; K 1 = 15; K 2 = 27; K 3 = 30; K 4 = 48; and K 5 = 70 Fig. 6: Simulator status before execution of data mapper function of v.90 digital modem Fig. 7: Simulator status after execution of data mapper function of v.90 digital modem Data mapper associated with the particular data frame interval 'i' selects PCM code word corresponding to the U-code (C i ) specified by output of modulus encoder (Ki) from the PCM code words table as shown in Appendix (A). After debugging the program, it was executed on simulator whose status is shown in Fig. 7. As the outputs of data mappers have been directed to stored at 'dma' BD00H to BD05H in the program, It is clear from the Data Memory window of simulator that data at these locations matches with the PCM code words of Appendix (A) corresponding to K i values taken as an example. This confirms the successfulness of the algorithm as well as corresponding assembly program designed during the present study. Implementation of data mappers on TMS320C50PQ57 DSP: As discussed earlier, a program in assembly language has been developed to implement data mapping function on TMS320C50PQ57 DSP. Universal PCM code words from 'pma' to 'dma' have been shifted with the help of block move instruction BLKP (block data move from program memory to data memory ) and RPT (repeat) instruction. In data mapping function, Auxiliary Resisters AR1 and AR2 have been utilized which also aids in performing query task given in the flow diagram discussed earlier. After debugging the program, it has been loaded into the DSP module with the help of assembler and linker programs. Program has been executed on the module by taking the same inputs K i (output of the modulus encoder) as assumed during simulation of the program. It is very clear that data stored at data address BD00 to BD05 exactly matches with the PCM universal codes corresponding to assumed K i, which confirms the reliable implementation of data mappers on DSP. Different implementation parameters corresponding to Data mapping functionality have been given in Table 1. Table 1: Summery of implementation parameters Data Memory Program Memory Program Execution Function Used (W) Used (W) Time (? s) Data Mappers 134 164 2.877 605

CONCLUSIONS The study has presented simulation of the Algorithm developed for data s mapper for the transmitter of V.90 Digital Modem. Implementation of Data Mappers on Digital signal processor has been carried out and different implementation parameters like data memory, memory used and programmed execution time worked determined. REFERENCES 1. Anu, A.G., 2001. Introduction to telecommunications. Thomson Asia Pvt. Ltd., Singapore. 2. Clarke, D.E.A. et al., 1998. Emerging Broadband Access Technologies. British Telecommunication Technol. J., 16: 187-195. 3. Reddy, V.U., 2001. Voice-band modems: a device to transmit data over telephone networks-part (I)- Basic principles of data transmission. Resonance, 6: 56-65. 4. Reddy, V.U., 2001. Voice-band modems: a device to transmit data over telephone networks-part (II)-Advanced ideas whichh made high data rates possible. Resonance, 6: 60-70. 5. Adams, F. et al., 1998. Today's access technologies. British Telecommunication Technol. J., 16: 21-33. 6. Georgia Tech Studies <http://www.gvu.gatech.edu/user_surveys/survey- 1998-10/graphs/technology/ q01.htm> 7. ITU-T Recommendation V.90, 1998. A digital and analog modem pair for use on the PSTN at data signaling rates of up to 56000 bits/s downstream and 33600 bits/s upstream. ITU-T V Series Recommendations. 8. Recommendation G.711, 1998. Pulse coded modulation (PCM) of voice frequencies. ITU-T G Series Recommendations. 9. Les Brown, 1998. PCM modem design: V.90 characteristics. Communication System Design Magazine. http://www.csdmag.com/main/9806fe2.htm 10. Steven, A.T., 2002. Contellation shaping, nonlinear precoding and trellis coding for voiceband telephone channel modems. Kluwer Academic Publishers, UK. 606