Reinforcement Machine Learning-based Improved protocol for Mobile Ad-Hoc Networks
Keywords:Mobile ad-hoc network, Reinforcement learning, K-Means Clustering, Machine Learning, Clustering, Ad-hoc on demand distance vector etc
Mobile Ad-Hoc Networks (MANETs) serve as a vital communication infrastructure in scenarios where fixed infrastructure is absent or impractical. Energy efficiency is a critical concern in such networks, given the inherent constraints of battery-powered mobile devices. This research introduces a groundbreaking approach—a Reinforcement Machine Learning-based Improved Energy Efficient AODV (Ad-Hoc On-Demand Distance Vector) Protocol (RML-EEAODV)—to address the pressing need for enhanced energy utilization in MANETs. RML-EEAODV combines the strengths of reinforcement machine learning with the AODV routing protocol to create an intelligent and adaptive energy-efficient routing solution. In the current situation, the most significant challenge in MANET is to reduce energy consumption and overhead of each node, and to give a better packet delivery ratio. Reinforcement Machine learning can be used to improve routing decisions of AODV protocol by encouraging nodes to updates a state information database of intermediate nodes along routes to destinations. state information is used for taking forwarding decision to find guaranteed QoS routes. Our proposed solution RML-EEAODV is highly efficient in terms of energy consumption, network overhead and Gives the acceptable level of packet delivery ratio.