Energy-Efficient Routing Protocol for Mobile Ad Hoc Networks Using Machine Learning
Abstract
Energy efficiency remains one of the most critical challenges in Mobile Ad Hoc Networks (MANETs) because network nodes operate using limited battery resources while communicating over dynamically changing wireless topologies. Frequent route discoveries, unstable links, high node mobility, and excessive control overhead significantly increase energy consumption, reducing network lifetime and degrading communication performance. Conventional routing protocols such as Ad hoc On-Demand Distance Vector (AODV), Dynamic Source Routing (DSR), Optimized Link State Routing (OLSR), and Destination-Sequenced Distance Vector (DSDV) primarily focus on route establishment and packet delivery without adequately considering residual battery power, node stability, traffic congestion, or future network conditions. Consequently, these protocols often suffer from premature node failures, uneven energy depletion, increased routing overhead, and reduced Quality of Service (QoS) under highly dynamic MANET environments. Early research between 2008 and 2018 introduced numerous energy-aware routing strategies, including residual-energy-based routing, cross-layer optimization, fuzzy logic, swarm intelligence, and reinforcement learning concepts, providing the foundation for intelligent routing mechanisms. This study proposes an Energy-Efficient Routing Protocol for Mobile Ad Hoc Networks Using Machine Learning (EERP-ML) that integrates energy-aware routing metrics with lightweight machine learning techniques to improve route selection and extend network lifetime. The proposed framework combines residual node energy, node mobility, link stability, hop count, traffic load, and historical routing information to predict optimal communication paths dynamically. Machine learning is employed to classify candidate routes according to their expected energy efficiency and reliability, allowing the routing protocol to adapt to changing network conditions while minimizing unnecessary route discoveries and packet retransmissions.
