ESRQ: An Efficient Secure Routing Method in Wireless Sensor Networks Based on Q-Learning

摘要

Some trust-based methods have been proposed for consideration by guaranteeing the security routing of wireless sensor networks (WSN) recently. However, Most of the current research work usually consumes a lot of energy resources and computing resources or only performs well for specific types of attacks. In this paper, we propose a lightweight and efficient security routing method by combining the trust mechanism with q-learning algorithm. The sensor node determines the credibility of the specific node autonomously and efficiently according to the performance of the behavior of the node. The results show that 1) this method provides a more secure routing mechanism and it can effectively deal with many kinds of attacks; and 2) the method is lightweight and reliable, it provides better energy/packet delivery tradeoff performance compared with other existing trust-based methods and can cope with different conditions.