DYNACLUSTERNET: A Dynamic Clustering Network‐Based Routing Optimization for Edge Computing

Abstract The Internet of Things has led to a surge in data generation and network complexity, especially in edge environments with dynamic topologies and moving objects. Traditional clustering methods in edge computing often fail to address these challenges, such as efficient data aggregation and computational management. DynaClusterNet, a novel framework, introduces three protocols: Adaptive Cluster‐Based Deployment Protocol (ACDP), Dynamic Algae Spider Protocol (DASP), and Deep Q Routing Protocol (DQRP). The ACDP uses Voronoi diagrams for optimal node deployment and cluster formation, while the DASP uses Artificial Algae and Black Widow Algorithms to dynamically select cluster heads and optimize data transmission. The DQRP uses deep reinforcement learning to determine efficient routing paths, adapting to environmental changes, node mobility, and evolving network topologies. DynaClusterNet significantly outperforms existing protocols in terms of end‐to‐end delay, energy consumption, and Packet Delivery Ratio. It ensures a robust, efficient, and adaptable network performance with a minimal end‐to‐end delay of approximately 0.05 s and significantly lower energy consumption profile than competing protocols.