Mobility Aware Playout Algorithm for Interactive Audio Streaming over Wireless Ad-Hoc Networks

摘要

Interactive audio streaming over mobile wireless ad-hoc networks is a difficult task since these networks exhibit challenging characteristics such as mobility of all nodes, limited end-to-end bandwidth, high error rate, etc. Interactive network applications require the reception of each media unit before its deadline in order to guarantee a continuous playback process. To satisfy this constraint, a de-jitter buffer is installed on receiver host in order to mitigate the effect of delay jitter on perceptual quality. To maximize mitigate the effect of delay jitter on perceptual quality. To maximize the perceived quality and to minimize the buffer space allocated by the playback audio process, the de-jitter delay must be adapted dynamically within an audio session. In this paper, we propose an adjustment playout algorithm for interactive audio applications over wireless ad-hoc networks. The proposed algorithm denoted MAPA (Mobility Aware Playout Algorithm) adjusts the playout delay based on both node mobility, which characterizes mobile ad-hoc networks, and talk-spurt, which is the intrinsic characteristic of a voice signal. The mobility detection mechanism is done at the receiver application layer. We assess the perceptual quality using an advanced evaluation method based on E-Model concepts and including the variable nature of impairment during an audio session. Simulation results show that the proposed algorithm outperforms significantly conventional playout algorithms specifically over a highly varying topology.