MPC-based Delay-Aware Fountain Codes for Live Video Streaming

The explosive demand for live video streaming over wireless networks is constantly calling for innovations in both video coding and wireless communities. The state-of-the-art answer to this challenge are sliding-window-based Delay-Aware Fountain (DAF) codes, which combine the channel-adaptive feature in rateless coding and the delay-aware feature in video coding. However, a high computational cost is incurred during the per-window optimization of sampling pattern, making DAF codes impractical for live streaming. To solve this issue, we propose a novel online algorithm for DAF codes. By integrating the Model Predictive Control (MPC) technique into DAF codes, a finite length of horizon is imposed on the optimization algorithm. As a result, the complexity is lowered to an affordable level such that real-time video encoding is supported. Two schemes are developed in this paper: (i) DAF-M, the MPC-based DAF codes, and (ii) DAF-O, the online variant of DAF-M based on video bit rate prediction. The advantages of both designs are validated through comprehensive experiments. The results of simulation experiments show that the decoding ratio of DAF-M is close to the global optimum in DAF codes, and higher than the other existing schemes; DAF-O outperforms the state-of-the-art live streaming algorithms.