A Novel Congestion Control Scheme using Firefly Algorithm Optimized Fuzzy-PID Controller in Wireless Sensor Network

Abstract

Wireless Sensor Networks (WSNs) consist of several sensor nodes, each of which may collect, receive and transmit data. In recent years, WSNs have emerged as essential technologies due to their ubiquity in applications such as the military, smartphones, disaster management, healthcare monitoring, and other surveillance systems. The inability to send data from the sensor node promptly and the impossibility of new data reaching the node's queue indicate of network congestion. The packet will be either discarded or delayed, which will cause more data loss, longer transmission delays, reduced network throughput, and lower network quality of service. To address this problem, this paper proposes an efficient and novel Firefly Algorithm-optimized Fuzzy-PID (FA-Fuzzy-PID) controller for congestion control in Wireless Sensor Networks (WSNs). The proposed control technique used a fuzzy control algorithm to overcome the standard PID controller's slow optimization parameter, low calculation accuracy, and limited adaptability. The Firefly Algorithm (FA) was used to optimize the PID parameters increment from the Fuzzy-PID controller. The proposed controller was designed in MATLAB and analyzed using Network Simulator 3. Simulation results from the proposed control method achieved 92.00% higher throughput, 44.49% lower packet loss rate, and 75.00% lower for both congestion level and queue size when compared to Cuckoo Search FuzzyPID (CFPID) and Particle Swarm Optimization-Gravitational Search Algorithm (PSOGSA). These improvements will allow regular data flow and improve the overall performance of WSNs. Based on these improvements, the proposed technique is an efficient control method for congestion control in the WSN because end-users will experience faster data transfer speeds and fewer network delays.