AI-Enhanced CRT-LEACH: A Novel Routing Protocol for Prolonging the Lifetime of Wireless Sensor Networks

Abstract

Wireless Sensor Networks (WSNs) are increasingly deployed in critical monitoring applications, ranging from environmental surveillance to industrial automation. However, their operational efficiency is frequently constrained by limited energy resources and suboptimal routing mechanisms. This study proposes an AI-enhanced CRT-LEACH (Chinese Remainder Theorem–Low Energy Adaptive Clustering Hierarchy) routing protocol designed to address the challenges of energy consumption, network lifetime, and performance degradation in dense WSN environments. The protocol integrates machine learning techniques, specifically a lightweight reinforcement learning algorithm, to dynamically optimize cluster head (CH) selection based on node residual energy, communication distance, and historical network behavior. The Chinese Remainder Theorem is incorporated to ensure fault-tolerant and deterministic scheduling of data transmission paths, minimizing redundant broadcasts and transmission collisions. Simulation experiments were conducted using NS-3 to compare the proposed AI-CRT-LEACH protocol against standard LEACH and CRT-LEACH variants. Results demonstrated a significant improvement in network lifetime, extending up to 47% longer than LEACH and also reduced average energy consumption per round. Additionally, the AI component exhibited superior adaptability to node failures and mobility, leading to enhanced packet delivery ratio and reduced latency. This hybrid approach illustrates the potential of combining algorithmic efficiency with intelligent decision-making to foster more resilient and sustainable WSN deployments, especially in resource-constrained environments.