The subband modulation: a joint power and rate allocation framework for subband image and video transmission

A new approach of reliable image and video transmission over noisy channels is proposed. For subband decomposed image and video, combined source coding and channel modulation design can achieve high compression efficiency and preferable quality. Further performance gain is obtained by multiresolution modulation as well as a bit-remapping scheme that assigns efficient mapping from each source codeword to channel modulation points. We show that the combined source coding and modulation design outperforms conventional approaches, which design source coding and modulation separately. A simple channel distortion approximation is derived by applying a bitremapping scheme, which allows the power allocation to be employed to further enhance the performance. Compared to the joint source and channel coding with a binary phase shift keying modulation system and fixed modulation with the one-to-one intelligent mapping system, the proposed system performs better in a middle-range signal-to-noise ratio and low channel bandwidth. The simulation is carried out on additive white Gaussian noise channels.