Joint Source-channel Coding Research Based On Real BCH

In a wireless multimedia communication system, multimedia data stream coding efficiency and error resilience are major contradiction. In the source coding, improving coding efficiency will reduce the streams’ anti-error performance; in the channel coding,improving streams’ error resilience will reduce the coding efficiency. The problem needs to be solved by source coding and channel coding is conflicting. Joint source-channel coding based on real BCH is an effective way to solve the above coding contradictory.The paper built the joint source-channel coding system based on real BCH. Joint channel is composed by quantization, a binary symmetric channel given the transition probability and inverse quantization. Joint channel is modeled by GBG( Gaussian Background Noise and Bernoulli Gaussian Impulse Noise). Joint source-channel coding system based on real BCH considers the quantization noise and channel noise. This article will study the optimal bit allocation of Lloyd-Max quantizer. A new threshold algorithm of real BCH error correction codes are found based on Bayesian hypothesis testing. Real BCH error correction codes’ error correction performance is analyzed under the new threshold algorithm.Matlab simulation results show that in different channel transition probability conditions, the image has good transmission performance in real BCH joint sourcechannel coding system. When the channel quality is better, the receiving side can obtain the higher the quality of reconstructed signal, with the channel quality deteriorates the quality of reconstructed signal decreases at the receiving side. But even if the channel quality is very bad, the quality of reconstructed signal at the receiving side can still be accepted. The new Threshold algorithm based on real BCH error correction code is simple and effective.