Modeling and Simulation of Modeling and Simulation of Turbo Encoder and Decoder for Turbo Encoder and Decoder for Wireless Communication Wireless Communication SystemsSystems

Sayantan Choudhury

Problem StatementProblem Statement

• Implementation of a turbo encoder and decoder

- Maximum A Posteriori (MAP) algorithm- Sliding window implementation

• Performance metrics : - Bit Error Rate (BER)- Signal to Noise Ratio (SNR)

• For a fixed BER, SNR requirement is reduced (Coding Gain)

Turbo EncoderTurbo Encoder

D D D

Interleaver

D D D

Information Sequencedk

Encoder 1

Encoder 2Zk

Xk

Yk

Turbo DecoderTurbo Decoder

DecoderDEC 1

Inter-Leaving

DecoderDEC2

Deinter-leaving

Deinter-leaving

Feedback Loop

Decoded Output

Zk

xk

yk

dk

Le2 (dk)

L2 (dk)

L1(dk)L2 (dk)

ImplementationImplementation

• SDF model of computation

• Use MAP algorithm for decoding – Advantage : Higher coding gain at low SNR – Disadvantage : More computationally and memory

intensive

• Enhancements– Sliding window

– MAX* approximation

Performance EvaluationPerformance Evaluation

Performance Factors and TradeoffsPerformance Factors and Tradeoffs

• Selection of algorithm• Constraint length and number of

iterations• Interleaver design • Trellis termination • Puncturing• Frame size

ConclusionsConclusions

• Successful implementation of Turbo Encoder and Decoder

• Written mainly in C• Works with TMS320C67 DSK• Well designed code should be very easy

to port to Advanced Design Systems (ADS)