Research On Channel Coding And Decoding Methods Based On Turbo Codes

Turbo codes, as soon as their appearance have caused the stir in information and code theorists due to their outstanding performance of closing with Shannon limit, and they obtain widespread research and application. They are determined as one of the third generation mobile communication system’s channel coding plans. The near Shannon performance of the Turbo codes is mainly due to their iterative decoding. But their complexities and decoding delay are large because of iterative decoding. So improving and simplifying the decoding algorithm are the one of the current research direction.After an introduction to the coding and decoding principles of Turbo codes and derivation of the decoding algorithms, the paper mainly studied the improved decoding algorithm and the iterative stopping criterion. Firstly, the paper simulated the design parameters on the bit error rate performance of Turbo codes such as the component coding, zero-dedect processing method, the length of interleaver, coding rate, interwoven device, the number of iteration and decoding algorithms with MATLAB. The reasonable parameters were chosen to enhance the performance of Turbo codes.Based on some research and analysis about the traditional Turbo decoding algorithms, in order to reduce decoding complexity and decoding delay, the improved iterative decoding algorithms such as improved sliding windows log max a posteriori (improved SW-Log-MAP) and improved combination decoding algorithm were presented. The application of sliding windows in improved SW-Log-MAP algorithm reduced the decoding delay, simplified the Log-MAP and made the amount of calculation greatly reduced; the improved combination decoding algorithm combinated the advantages of soft-output Viterbi algorithm (SOVA) and Log-MAP, at the same time, the algorithm achieved the Log-MAP decoding performance and reduced decoding complexity. The results showed that the improved algorithms could reduce complexity while the performance approximated to Log-MAP. To reduce decoding delay, a new improved iterative stopping criterion was brought up. The criterion utilized the statistical characters of logarithm likelihood ratios from output of the two constituent decoders. The simulations showed that the criterion could reduce the average number of decoding iteration while the bit error rate performance was almost not degraded.Finally, Turbo codes were applied to orthogonal frequency division multiplexing (OFDM) system as channel coding, the simulation results showed that the system could improve anti-interference and resist Rayleigh fading characteristics, thus the transmission speed and quality were increased.