FOA-IT2FLS+QL: An Adaptive Energy-Efficient Clustering and Routing Protocol for IoT-enabled Wireless Sensor Networks

Abstract

This research paper suggests a hybrid energy-efficient clustering and routing method inspired by the Fruit Fly Optimization Algorithm (FOA) and Interval-Type-2 Fuzzy Logic System (IT2FLS) and Q-Learning for Internet of Things (IoT) - Wireless Sensor Networks (WSN). First, FOA is applied to decide the best set of Cluster Heads (CHs) and then an IT2FLS is used to judge the candidate's fitness in the uncertain network based on distance to BS (Base Station), Residual Energy and node degree. Optimized CHs are then used for a Q-Learning routing algorithm, which is able to create secure multi-hop communication paths. The routing mechanism adds residual energy and distance progress to the reward function to enhance reliability, and prevent the formation of hotspot. The IT2FLS optimizes cluster formation and routing jointly to achieve the best results. The proposed technique compared with well-known algorithms; LEACH, DEEC and BOA+ACO and performs best with regard to energy efficiency, packet delivery, throughput and network lifetime.