Efficient wireless sensor networks scheduling scheme: Game theoretic analysis and algorithm

Energy efficiency is the core issue in wireless sensor network´s scheduling algorithm design, which calls up wide attention from researchers. However, few of the previous works notice the constraints on sensors caused by limited buffer size. It challenges the system´s meeting deadlines performance. To alleviate this concern, we propose a scheduling policy: the least energy and free storage remaining first, which extends wireless sensor network´s lifetime and provides a high real-time service´s quality. To make our strategy more practical, we consider the user´s non-cooperative behavior out of self-interests. It increases the system´s packet lost in buffer overflow and reduces the energy efficiency. Based on a mixed strategy game model, we show that the non-cooperative game converges to an inefficient Nash Equilibrium, where the spectrum resource is significantly wasted. In order to eliminate user´s selfish manner, we design a punishment scheme via a repeated game, which considers the time-variant character of network and can detect non-cooperative action precisely and rapidly.