Joint source channel coding for wavelet-compressed image transmissions using packet-level FEC

This thesis proposes a joint source channel coding (JSCC) method for wavelet-compressed image transmission in packet networks where data loss during transmission is characterized by packet loss rate (PLR). To combat packet loss within a fixed delay required by real-time services such as real-time video, packet-level forward error correction (FEC) is deployed which adds redundancy into the data stream. The objective of this thesis balances the compression rate in source coding and the coding (redundancy) rate in the channel code so as to minimize the mean squared error (MSE) metric of the reconstructed image for transmission over a channel with fixed capacity.; The joint source channel coding technique proposed in this thesis improves the quality of image transmission over imperfect channels characterized by high PLR and fixed capacity. (Abstract shortened by UMI.)...