Cognitive Fish Swarm Optimization for Multi-Objective Routing in IoT-based Wireless Sensor Networks utilized in Greenhouse Agriculture

Abstract

This research presents the working mechanism of Cognitive Fish Swarm Optimization (CFSO) for multi-objective routing and channel selection in Internet of Things (IoT)-based Wireless Sensor Networks (IWSNs). CFSO is inspired by the collective intelligence and cooperation observed in fish swarms. The model involves three main components: perception, cognition, and behavior. Each fish in the swarm perceives the network conditions by gathering information from its surrounding environment, including signal strength, channel availability, and network congestion. The fish then utilizes its cognitive abilities to evaluate different routing paths and channel options based on specific objectives, namely energy efficiency, packet delivery ratio, and delay. This evaluation process involves analyzing historical information and utilizing heuristics to create notified results. Each fish adapts its behavior by adjusting its movement pattern and selecting optimal routing paths and channels. This adaptive behavior is critical for achieving reliable and efficient data transmission in IWSNs. The fish swarm balances exploration and exploitation strategies to search for optimal solutions comprehensively. Exploration allows for discovering new paths and channels, while exploitation focuses on refining the best-known solutions. The efficiency of the CFSO method in enhancing data transmission efficiency in greenhouse agriculture applications was validated through extensive simulations in the NS-3 network simulation framework. The findings suggest that the CFSO method is a promising technique for addressing routing and channel selection challenges in IWSN by leveraging the collective intelligence of fish swarms. The CFSO model portrayed a superior throughput and Network Lifetime (NLT) values of 71.34% and 77.20%, respectively, significantly outpacing SSEER and CRP across overall node counts.