Self-learning scheduling approach for wireless sensor network

Energy efficiency is one of key issues of wireless sensor network (WSN). In this paper, we propose a self-learning scheduling approach (SSA) to reduce energy consumption for wireless sensor network (WSN). This approach integrates sleep scheduling together with packet transmission scheduling to reduce energy consumption. It enables nodes to learn continuous transmission parameters and sleep parameter through interacting with the WSN. The continuous value of transmission parameter is achieved by our extension of Q-learning method, and the value of sleep parameter can be calculated from the transmission parameter. We valid this approach in a MAC protocol and compare some network performances between the SSA and SMAC protocol. The simulation results show that our SSA performs much better than SMAC protocol in these QoS metrics.