Joint Source/Channel Coding for JPEG2000 Codestream Transmission over Fading Channels

Abstract

The embedded codestream structures and a variety of progressive image compression methods provided by JPEG2000 make it attractive for image transmission purposes. A three-level, hybrid optimization scheme is proposed for JPEG2000 compressed images over a correlated block fading channel with a cyclic redundancy check code (CRC) and a rate-compatible punctured convolution (RCPC) code. It combines forward error correction (FEC) and JPEG2000 error resilience (ER) mechanisms to minimize the end-to-end image distortion according to channel conditions. The proposed scheme employed a finite-state Markov channel model to simulate the channel. It has the complexity less than O (R2) for a special JPEG2000 codestream, where R is the number of available channel code rates. Simulation results show a significant improvement about 0.7-1.3 dB over the based two-level scheme in quality of received images