Correlated Node Behavior Modeling Using Fuzzy In Wireless Sensor Network

ABSTRACT Performance of Wireless Sensor Network depends highly upon the cooperation among the nodes behavior in transmission of packet data, messages and route discovery. Over open medium environment, nodes are free to move and may change their behavior arbitrarily. In the presence of misbehavior node, it may instigate the spreading of abnormality among the neighbour nodes. This resulted to a correlated node behavior and gives significant impact towards network performance. Fuzzy logic system is proposed as an alternative for crisp values which can effectively enhance the computational complexity of event detection. Therefore, this research aims to formulate correlated node behavior detection using fuzzy system. The formulation of correlated node behavior is determined from the current status of node behaviour. Then, the comparative correlation techniques namely spatial and attribute correlation are implemented to address the correlated degree determination. The effectiveness of model is established via MATLAB simulation. The result of this study shows that both the attribute and spatial correlation are significant to determine the correlated degree of node behaviour. The research also analysed the correlated fuzzy function in terms of network performance parameters. Hence, this research provides a fundamental guideline for network designer to understand the fault-tolerance in network topology.