Improving TICK Efficiency by GA-based Fuzzy Membership Functions Optimization in Wireless Sensor Networks

In wireless sensor networks (WSNs), the energy and computational capacity of the sensor nodes are limited and they are deployed in a hostile environment. An adversary can easily compromise a sensor node and injects false data into the network. This injection of false data drastically depletes the energy of the nodes designated on the route for forwarding the source data. Many filtering methods have been proposed to tackle the security issues but they also increase the communication cost by exchanging the control keying messages which consumes more energy with in the network. Fuzzy logic for TICK was proposed to reduce the communication cost and increase the energy efficiency of the en-route nodes by selecting the re-encrypting nodes efficiently and avoiding the exchange of control keying messages in the network. Membership functions in the fuzzy system need to be optimized in order to use the fuzzy inferencing more efficiently. We propose an optimized fuzzy membership functions method by using genetic algorithm. The genetic algorithm determines the optimal membership functions which help in selecting the most favourite re-encrypting nodes in TICK based wireless sensor network. Simulation results show that the proposed method optimizes the membership functions and achieves better energy conservation at sensor nodes.