Energy and Surveillance Quality Sensitive Scheduling Scheme for Composite Event Detection in Wireless Sensor Networks

Abstract

Event surveillance especially the composite event surveillance is a crucial task for wireless sensor networks applications. However, sensor nodes are vulnerable by harsh environmental conditions, which cause sensor nodes fallibility and affect the quality of surveillance. In order to improve the surveillance quality of composite event, one straightforward method is to increase the number of wake-up nodes in event surveillance, but this result in an immense consumption of precious energy. This paper focuses on improving surveillance quality of composite event in an energy-efficient manner. We use a three-layer statistically reliability model (StaR) to quantify the surveillance quality of composite event. Given the detection quality parameter and a k-out-of-n voting strategy, StaR-based scheduling scheme can dynamically schedule the nodes' states within a composite event detection group, i.e., inactivate low detection reliability node and activate high detection reliability node, to satisfy the user-defined composite event detection quality and realize energy conservation meanwhile.