Joint Source And Channel Coding For Image Transmission Using Turbo Codes

With the progress in the fields of computer, telecommunication and multimedia, etc., image communication, an essential component of multimedia communications, is studied extensively. The traditional communication system is based on Shannon's separation principle, which source coding and channel coding is operated dividedly. But the separation approach often leads to low efficiency especially for image transmission. Real-world systems benefit through joint design of source and channel coders, given knowledge of the channel. This thesis proposes a joint source and channel coding scheme using turbo codes for image transmission, which takes advantage of the available reliability information (generated by the Turbo-code decoder) to optimized the quantizer and reconstructor, improving the quality of the reconstructed signal. The following algorithms have been presented and tested in the thesis:In this thesis, we review the joint source and channel coding firstly, introducing fundamental theories, design methodologies, and practical applications.Secondly, a novel variable-rate image compress method, named as CMCVQ (Characteristic Values-Match Correlation Vector Quantization), is presented. Before encoding, four characteristic values of each codeword are computed, and then four corresponding sorted codebooks are obtained according to the ascending order of the four characteristic values of codewords respectively. During the encoding, high correlation of the adjacent image blocks and characteristic values of the current processing vector and codewords are considered to encode the current processing vector, which not only have lower bit rate, but also can be operated in shorter time.Thirdly, the principle of turbo-code coder and decoder are presented. The turbo-code decoder produces bit available reliability information expressed as a log likelihood ratio, which includes the channel information. We apply the information to the optimization of quantizer and reconstructor, and use soft reconstruction rule to improve the performance of the system. Based on the principle of minimizing mean square distortion, several channel models is presented to design different quantizer and reconstructor pair, and corresponding arithmetic...