Multi-Stream Rate Adaptation using Scalable Video Coding with Medium Grain Scalability

Multiple video streaming in a shared channel with constant bandwidth requires rate adaptation in order to optimize the overall qual- ity. In this paper we propose a multi-stream rate adaptation frame- work with reference to the scalable video coding (SVC) extension of the H.264/AVC standard with medium grain scalability (MGS) and qual- ity layer (QL). We first provide a general discrete multi-objective prob- lem formulation with the aim to maximize the sum of assigned rates while minimizing the differences among distortions under a total bit- rate constraint. A single-objective problem formulation is then derived by applying a continuous relaxation to the problem. We also propose a simplified continuous semi-analytical model that accurately estimates the rate-distortion relationship and allows us to derive an optimal and low-complexity procedure to solve the relaxed problem. The numerical results show the goodness of our framework in terms of error gap between the relaxed and its related discrete solutions, the significant performance improvement with respect to an equal-rate adaptation scheme, and the lower complexity with respect to a sub-optimal algorithm proposed in the literature.