Energy-efficient routing in mobile IoT networks: A reinforcement learning approach for optimal hub placement and data transmission

Abstract

The rapid expansion of the Internet of Things (IoT) has significantly increased energy consumption and affected network efficiency, particularly in mobile IoT networks where frequent hub movement accelerates energy depletion. This study proposes a machine learning-based approach for optimal hub placement and routing to enhance energy efficiency. Using reinforcement learning (RL) with a double Q-learning algorithm, the method involves data collection, IoT user clustering, priority setting based on urgency, data size, and energy levels, and training a deep RL network for efficient decision-making. The goal is optimal resource utilization with minimal effort and time investment. Performance metrics such as energy consumption, latency, and throughput evaluate the effectiveness of the proposed method. By reducing energy usage in mobile IoT networks, this approach extends device lifespan and promotes sustainability. Integrating advanced machine learning with IoT network management, the study offers a scalable and reliable solution, paving the way for future advancements in wireless communication technology.