Power-Rate-Distortion Optimized Resource-Scalable Low-Complexity Video Coding in Wireless Multimedia Sensor Networks

Wireless multimedia sensor networks (WMSNs) have been potentially applicable for several emerging applications. However, the available resources, i.e., power and rate, of visual sensors in a WMSN are very limited. Hence, it is important but challenging to achieve efficient resource allocation and optimal video compression while maximizing the overall network lifetime. In this paper, a power-rate-distortion (PRD) optimized resource-scalable low-complexity multiview video encoding scheme is proposed. In our video encoder, both the temporal and interview information can be efficiently exploited based on the comparisons of extracted media hashes without performing motion and disparity estimations, which are known to be time-consuming. We present a PRD model to characterize the relationship between the available resources and the RD performance of our encoder. More specifically, an RD function in terms of the percentages for different coding modes of blocks and the target bit rate under the available resource constraints is derived for optimal coding mode decision. Analytic results are provided to verify the resource scalability and accuracy of our PRD model, which can provide a theoretical guideline for performance optimization under limited resource constraints. Both the analytic and simulation results have shown the applicability of our video coding scheme for WMSNs.