Frame replenishment coding over noisy channels.

Abstract

In this thesis, we compare the performances of three frame replenishment coding schemes for noiseless and noisy channel transmission. For the noisy channel case, we propose a combined source-channel coding scheme which provides error protection in order to improve the performance. The three frame replenishment coding techniques are: basic frame replenishment coding (BFR), label replenishment coding using vector quantization (LRVQ) and codeword replenishment coding using vector quantization (CWRVQ). Our simulation results demonstrate that for a noiseless channel, the LRVQ and CWRVQ techniques provide better performance, and tolerance to motion in the picture, compared to BFR technique. However, in the presence of channel noise LRVQ and CWRVQ techniques exhibit high sensitivities, resulting in poor performance. We propose to use a combined source-channel coding technique with three different error protection schemes where the bit rates for the source code and channel code are adjusted so as to minimize the mean square error. Simulation results demonstrate that the three error protection schemes provide a significant improvement in performance for LRVQ and CWRVQ without sacrificing transmission bandwidth. (Abstract shortened by UMI.)