Optimizing Video Transmission over Wireless Multimedia Sensor Networks

This paper proposes an efficient packet scheduling scheme for minimizing perceived video distortion and power consumption over multipath wireless multimedia sensor networks. The proposed scheduling algorithm is responsible for transmitting the video packets over selected multipaths according to their importance (high priority packets over high bandwidth paths). Moreover, whenever the aggregate transmission rate available at the network cannot support the required transmission rate due to high packet loss probability, the scheduling algorithm can selectively drop combinations of video packets prior to transmission to adapt the rate of the sender to the limitations of the wireless channel capacity. The video packets that will be dropped prior to transmission are selected based on a recursive distortion prediction model that has been customized for H.264/AVC coding scheme and counts for isolated losses, burst of losses and losses separated with small lag. Two scheduling algorithms are proposed. The baseline scheme utilizes a novel distortion prediction model and decides upon which packet can be dropped prior to transmission based on the packets impact on the video distortion. Moreover, the power aware packet scheduling is an extension of the baseline capable of estimating the power that will be consumed by each node during the transmission; hence it can control the power consumption by selectively drop the least significant video packets prior to transmission. Simulation results indicate the efficiency of the proposed scheduling scheme in terms of received video distortion (PSNR) and power consumption.