Iterative decoding for error resilient wireless data transmission

Abstract

Both turbo codes and LDPC codes form two new classes of codes that offer energy efficiencies close to theoretical limit predicted by Claude Shannon. The features of turbo codes include parallel code catenation, recursive convolutional encoders, punctured convolutional codes and an associated decoding algorithm. The features of LDPC codes include code construction, encoding algorithm, and an associated decoding algorithm. This dissertation specifically describes the process of encoding and decoding for both turbo and LDPC codes and demonstrates the performance comparison between theses two codes in terms of some performance factors. In addition, a more general discussion of iterative decoding is presented. One significant contribution of this dissertation is a study of some major performance factors that intensely contribute in the performance of both turbo codes and LDPC codes. These include Bit Error Rate, latency, code rate and computational resources. Simulation results show the performance of turbo codes and LDPC codes under different performance factors.