Perceptual quality based error control for scalable on-demand streaming in next-generation wireless networks

Wireless video streaming is anticipated to be one of the key technologies for many wireless multimedia applications. However, due to the limited bandwidth of wireless networks, it is challenging to provide live or on-demand streaming over wireless networks, especially on-demand streaming, which consumes huge bandwidth. To cope with this problem, many periodic video broadcasting protocols have been proposed for on-demand streaming. However, little research has been done on the error control for these periodic broadcasting protocols. In this paper, we utilize forward error correction (FEC) to develop a quality-adaptive error control scheme for broadcast-based on-demand streaming in next-generation network/IP multimedia subsystem (NGN/IMS) architecture. We first develop an analytic performance model for the error control problem. Then, we develop an efficient error control scheme, called Quality-Based Error Control (QBEC). The main merit of QBEC is that it has low computation complexity and it can find a near-optimal allocation of FEC parity bytes among all streams to maximize the overall quality perceived by all users. Our simulation results indicate that QBEC is an effective and efficient error control scheme for wireless on-demand streaming.