An Integrated Approach to Sink and Sensor Role Selection in Wireless Sensor Networks: Using Dynamic Programming
Yang Liu, Yi-Ju Zhan and Jing Chen
In this paper we propose a dynamic programming approach named IRSDP (Integrated sink and sensor Role Selection using Dynamic Programming). IRSDP can better balance energy depletion among the nodes in wireless sensor networks (WSNs) to prolong network lifetime. IRSDP is designed for applications where the entire area needs to be monitored over the course of network lifetime. First, we define a novel role selection cost structure that incorporates coverage overlap to avoid critical sensor selection and “maximum connecting” sink selection cost that considers the least-residual-energy sink’s neighbor sensors. Secondly, we propose a new scalable mathematical model, using dynamic programming to compute selection schedules. Finally, we discuss the implementation issues of IRSDP. Analysis and simulations prove that IRSDP achieves significantly better network performance than Non-Opt approach and DAPR.
Keywords: Wireless sensor networks, dynamic programming, sink selection, sensor role scheduling