A scheduling strategy for P2P-TV systems using Scalable Video Coding

The adoption of scalable encoders can improve P2P-TV performance in heterogeneous scenarios, where peers with different capabilities coexist. In fact, each peer could dynamically adapt the resolution of the audio/video signals to its computation capabilities and available communication resources. Despite these clear benefits, casting a scalable encoder in a P2P-TV system requires sophisticated scheduling strategies in order to maximize user satisfaction. In this paper, we propose a novel chunk scheduler for layered video designed on the basis of control theory. It is able to significantly limit base chunk losses at the expenses of enhancement ones, thus maximizing overall video quality. Simulation results, conducted using a real-encoded video sequence and variable size chunks, have clearly shown the effectiveness of the algorithm, able to reach near-optimal performances.