Power Consumption Analysis for Distributed Video Sensors in Machine-to-Machine Networks

Among different sensors used in machine-to-machine (M2M) networks, video sensors can provide the richest information. However, much higher bandwidth and power consumption limit the feasibility of wide deployment. To integrate visual data acquisition ability into M2M networks, the design of low-power distributed video sensors is the key. This paper presents the power analysis of distributed video sensors. The power profile of a distributed video sensor node is first shown from measurement results, followed by detailed discussion of video coding engine selection, where the state-of-the-art H.264/AVC coding engine is compared with distributed video coding engine. Moreover, the role of video content analysis is addressed. This paper can provide a fundamental basis for future distributed video sensor design.