Progressive Image Communication over Binary Channels with Additive Bursty Noise

A progressive method for transmission of images over a bursty noise hannel is presented. It is based on dis rete wavelet transform (DWT) oding and hannel-optimized s alar quantization. The main advantage of the proposed system is that it exploits the hannel memory and hen e has superior performan e over a similar s heme designed for the equivalent memoryless hannel through the use of hannel interleaving. In fa t, the performan e of the proposed system improves as the noise be omes more orrelated, at a xed bit error rate. Comparisons are made with other alternatives whi h employ independent sour e and hannel oding over the fully interleaved hannel at various bit rates and bit error rates. It is shown that the proposed method outperforms these substantially more omplex systems for the whole range of onsidered bit rates and for a wide range of hannel onditions.