Maximum lifetime effective-sensing partial target-coverage in camera sensor networks

Due to the emerging multimedia applications, camera sensor network is getting more spotlight these days. The visual sensors adopted to implement camera sensor network are a kind of directional sensors, which are well-investigated in the context of directional sensor network in the literature. However, the design and implementation of camera sensor network pose additional challenges due to unique properties of visual sensors. In this paper, we consider an interesting sleep-wakeup scheduling problem in camera sensor network, namely the maximum lifetime effective-sensing partial target-coverage (ML-EPT) problem. Given a set of targets and a set of camera sensors with non-uniform remaining energy-level, the goal of ML-EPT is to compute a schedule of the camera sensors such that the continuous time duration, during which a partial coverage-level requirement over the targets is met, is maximized. We propose a new heuristic algorithm, namely the MEASURE-and-SLICE (MaS) algorithm, which are based on two sub procedures, MEASURE and SLICE. Our simulation result shows that MaS dominates its existing alternative on average, and is close to optimal in some restricted cases.