Fuzzy Based Energy Efficient Routing for IoT: Traffic Delay Optimization

Abstract

Energy conservation and optimized traffic delay are crucial challenges in Internet of Things (IoT) systems, particularly in wireless sensor networks (WSNs). This study presents a novel approach to address these issues through a fuzzy‐based routing protocol. Traditional methods often struggle to effectively enhance routing protocols with optimized packet size control. Researchers have proposed a combination of machine learning and evolutionary techniques to overcome this limitation and enhance energy efficiency in WSNs. The new approach called MOSPFNN (Multi‐Objective Spider Prey‐localized Fuzzy Neural Network) leverages fuzzy logic control (FLC) to select optimal paths for traffic‐aware multipath routing. Additionally, a new meta‐heuristic algorithm of SP optimizer is employed to develop an energy‐efficient spatial routing protocol with superior convergence and minimal local optima. The proposed protocol incorporates optimal FNN for congestion monitoring and traffic‐aware routing. Simulation results validate that new protocol outperforms existing methods in terms of average end‐to‐end delay and packet delivery ratio (PDR). Using MOSPFNN, traffic with different priority levels can achieve a successful PDR rate of over 92% and network lifetime of above 2% than previous network within a reasonable timeframe. This research contributes to advancing energy‐efficient and traffic delay‐aware routing protocols in IoT systems.