Density evolution analysis for joint iterative decoding of wireless progressive image transmission

In this paper, joint source-channel decoding (JSCD) methods are designed for JPEG2000 progressive image transmission over both AWGN and flat fading channels. A regular low-density parity-check (LDPC) code is selected as the error correcting code. Error resilience mode is used in the JPEG2000 codec. A framework for the density evolution analysis is given to show its validation for the proposed JSCD methods. A general modification algorithm that utilises feedback information from the channel decoder is proposed to update the log-likelihood ratio (LLR) of partial codeword bits. Simulation results show that the JSCD design accelerates the iterative LDPC decoding procedure while the overall performance in terms of the PSNR for reconstructed images outperforms the non-joint decoding approach.