A Novel lightweight secure routing mechanism using Deep Reinforcement Learning for Body Area Sensor Networks

Smart healthcare technologies in the cutting-edge fields of the Internet of Things (IoT), Internet of Medical Things (IoMT), and engineering technology provide smart healthcare solutions using Body Area Sensor Networks (BASN). However, the efficient routing of information and information security are critical aspects of medical sensor networks (MSN) because of the sensitive nature of health data and the potential risks associated with unauthorized access, manipulation, or disclosure. Ensuring efficient routing with the security of medical sensor networks involves various measures and considerations including confidentiality, integrity, and availability. This study deals with an Artificial Intelligence (AI)-based lightweight secured routing mechanism using Deep Reinforcement Learning methodology. The proposed methodology is a reliable reinforcement-learning-based routing mechanism (RRR); hence, it was named R3. The prominence of this work includes a novel reliable reinforced routing mechanism, optimized decision-making by the proposed AI methodology, optimized routing of data, utilization of the least possible energy for data transmission, high reliability by integrating trust in the algorithm, and a faster shortest path algorithm with fewer than four hops to yield the shortest path for various traffic rates. The performance of the methodology was intensively evaluated against several potential attacks. The proposed R3-MedNet methodology was compared with four traditional techniques: (distributed energy-efficient clustering algorithm (DEEC), Stable Election Protocol (SEP), LEACH (low-energy adaptive clustering hierarchy (LEACH), and HEED (Hybrid Energy-Efficient Distributed clustering). Significant variations were observed in the energy levels of the nodes during various epoch stages. Simulation results prove that the proposed R3-MedNet methodology is resilient to various attacks.