A Realistic Coverage Model with Backup Set Computation for Wireless Video Sensor Network

Abstract

Wireless Video Sensor Network (WVSN) are gaining increasing popularity due availability of low-cost CMOS camera and miniaturization of hardware. For many applications, it is difficult to have a pre-engineered deployment of video camera sensors which leads to redundancy. Owing to sectored coverage and random deployment, it becomes challenging to model video sensor coverage to find redundancy and suppress redundant video transmission. Several efforts have been made to model coverage redundancy considering 2-dimensional coverage. Field of View (FoV) of the camera sensor is in 3-dimensions, thus it is very difficult to model the coverage and identify overlap area for realistic camera. 3-dimensional coverage is largely an unexplored problem. In this paper, a realistic 3-dimensional pyramid camera coverage is assumed and backup set of nodes are computed. Backup set of a node is a set of video sensor nodes that collectively covers coverage area of the node under consideration. The approach presented in the paper identifies minimal cardinality set of backup nodes that can be used to adaptively duty cycle the video capture and transmission. The result shows that the number of nodes required to remain active to cover the sensor field is reduced and in turn average energy consumption of the network also reduces.