A Machine Learning-Based Routing Protocol for 6G-Internet of Things network: Performance Evaluation

The integration of sixth-generation (6G) technology and the Internet of Things (IoT) is poised to revolutionize data management and technological interaction by providing unprecedented data transfer rates, ultra-low latency, and enhanced connectivity. This paper addresses the challenges of efficient, reliable, and adaptive data transmission in 6G-enabled IoT networks through a novel Machine Learning-Based Routing Protocol. Utilizing Deep Q-Networks (DQN), the proposed protocol optimizes routing by continuously adapting to real-time network states, ensuring Ultra-Reliable Low-Latency Communication (URLLC), energy efficiency, and real-time adaptability. Traditional routing protocols often fall short due to the complex and dynamic nature of IoT environments, necessitating innovative approaches to balance factors such as energy consumption, latency, and reliability. Our proposed method involves several key steps: initialization of the network, data collection and preprocessing, training of the machine learning model, route discovery and optimization, and data transmission. The reward function in DQN is designed to maximize energy efficiency, reduce latency, and enhance reliability. Experimental results, conducted using the Network Simulator 3 (NS3) and compared with existing methodologies such as Deep Deterministic Policy Gradient with Prioritized Experience Replay (DDPG-PER) and Speed-optimized Long Short-Term Memory (SP-LSTM), demonstrate the superiority of our approach.