Iterative source-channel decoding for robust image transmission
Wei Xiang, Steven S. Pietrobon, S.A. Barbulescu · University of Southern Queensland ePrints (University of Southern Queensland) · 2003
In this paper we investigate iterative source-channel decoding for robust JPEG coded images transmission over noisy channels. Huffman codes used as the variable-length coding scheme in JPEG coding can be represented by an irregular trellis structure proposed by Balakirsky (1997). State transition probabilities can be derived from the irregular trellis and can be used as a priori information to help iterative decoding between channel and source a posterJori probability (APP) decoders. Iterative decoding of JPEG coded images does not perform well due to the poor distance property of the original JPEG Huffman codes. In this paper, we proposed a symmetric reversible variable length code with free distance df 2 which can dramatically improve the system performance when iterative decoding is adopted. A maximum coding gain of 4 dB was observed from the simulation results. Subjective results in terms of reconstructed images were also presented.