Performance evaluation of fuzzy logic-based congestion optimisation approach for sensor networks

Abstract

Congestion in WSN leads to excessive consumption of energy, which needs to be controlled to improve the system life time. Congestion estimation has always been an important issue in wireless sensor networks for efficient information processing. In the recent past some researchers have used fuzzy logic for congestion estimation and shown that fuzzy logic is a potential tool for detection and estimation of congestion in wireless sensor networks. In this work we have used four different fuzzy systems for congestion estimation and evaluated their performances. Through these four fuzzy systems we will estimate the number of dropped packets with respect to time. In addition to this we have also found the average number of packet dropped for time interval 0 to 100 seconds, with respect to number of nodes in a certain cluster. Three important parameters affecting information processing and congestion in wireless sensor networks are identified as packet forwarding ratio, delay and validity. Simulation results of these fuzzy systems show that validity and delay-based system gives best performance and minimises packet drops to a significant amount during information processing in wireless sensor networks.