Stochastic rate-control of video coders for wireless channels

Abstract

We study the transmission of real-time video over wireless channels, proposing a formulation of the problem that includes a priori stochastic models for both source and channel. Using techniques of stochastic dynamic programming, we obtain offline optimal policies for each system state that minimize the average expected frame distortion. The online complexity of our approach is only that required to identify the state of the system (source and channel). The state of the channel is obtained based on the ARQ error-control mechanism, and the source state is computed as a complexity measurement on each incoming frame. Simulation results based on this new approach are provided and compared to other proposed rate-control strategies. They show how our model-based optimal policies require negligible on-line computational cost while providing a transmission quality comparable to that achieved with more complex deterministic dynamic programming techniques, and better than for simpler algorithms such as TMN.