Abstract
This chapter deals with iterative joint source channel decoding and its application to image transmission. First, the problem of transmitting a correlated gaussian source over an AWGN channel is considered. The joint decoding is implemented by the Baum Welch algorithm estimating the source statistics. Iterations between the MAP channel decoder and the source decoder are made to improve the global decoder performance. This decoding scheme is then applied to an image transmission system, based on a wavelet decomposition of the source image followed by a DPCM coding of the lowest frequency subband and a SPIHT coding of high frequency subbands. Simulation results show that a significant performance gain is obtained with iterative joint source channel decoding, compared to a classical decoding, in case of a correlated gaussian source and also in case of image transmission.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
