PAS: Prediction-based Adaptive Sleeping for Diffusion Stimulus Monitoring Sensor Networks

Energy efficiency has proven to be a dominating factor for the lifetime of WSN surveillance systems. Intensive studies have been done to provide energy efficient power management mechanisms, but few directly address the issues specifically for the sensor networks that monitoring Diffusion Stimulus (DS), a scenario that is the most common for environment monitoring applications. In this paper, we present our Prediction-based Adaptive Sleeping (PAS) solution that is designed directly for DS monitoring WSNs. The proposed PAS takes into account the characteristics of the DS spreading process and optimizes the sensors' sleeping schedules accordingly. The sensors near the DS boundary stay awake to timely capture the possible arrival of stimulus while the sensors far from the DS boundary keep in the energy-saving mode. The novelty design of PAS strikes a good balance between response time and energy conservation. Our simulation results also prove that the proposed PAS can greatly reduce the energy utilizations without noticeably sacrificing the system's performance.