Joint source and channel coding for image transmission over noisy channels with feedback

We propose solutions to the rate allocation problem of progressive image transmission over noisy channels with feedback. The scenario we are considering is the feedback constraint on each packet, which means the number of feedback bits on each packet is upper bounded by a fixed given value. Both rate-optimal and distortion-optimal solutions are investigated. Simulation results show that the proposed solutions outperform the previously published results and can closely achieve the performance of ARQ scheme with infinite number of feedback bits as the number of feedback bits increases. Although the simulations are performed using JPEG2000 source coder and rate-compatible LDPC channel codes, the solutions presented in this thesis are applicable to any progressive source coding scheme, any family of rate-compatible channel codes, and any channel model as long as the distortion-rate function of the source and the error probabilities of the channel codes are available.