Distortion optimized scheduling and QoS driven prioritization of video streams over WLAN

This paper presents a distortion optimization scheme. It involves a packet scheduling algorithm, based on a novel distortion prediction model, and a prioritization mechanism that categorizes the transmitted packets according to their video distortion importance to different priority classes. In the proposed scheme the video streams have been encoded based on the H.264/AVC codec with its multireference frame capability. A distortion prediction model, designed to capture the multi-referenced frame coding characteristic, is used to predetermine the distortion importance of each video frame in the stream. This information is prestored in each packet header. A scheduling algorithm will select which individual packet or combinations of packets will be drop based on the available transmission rate of the communication network, which in this case is a 802.11e WLAN. The network has been configured to differentiate the transmitted packets according to their distortion impact. Significant video packets are appointed to high priority access categories while least significant packets to lower categories. Simulation results show the improvement in the received video quality between the proposed scheduling scheme and an additive-distortion scheduling approach that omits the correlations among neighboring video frames. Moreover the simulation results show the improvement on the received video quality when the prioritization mechanism is implemented in addition to the packet scheduling algorithm.