Enhancing AODV Routing with Machine Learning for Intelligent IoT Path Selection

In the growing Internet of Things (IoT) landscape, the massive data generated by interconnected devices places significant demands on network infrastructure, requiring efficient routing protocols. The Ad Hoc On-Demand Distance Vector (AODV) protocol, however, faces frequent route failures, disrupting data transmission and increasing overhead. To address these issues, this study modifies the AODV route selection mechanism by integrating a decision tree algorithm. This approach evaluates multiple paths, selecting the most optimal route to enhance routing stability. Incorporating decision trees within the AODV framework improves route reliability and reduces disruptions. Simulations show this modification significantly boosts network performance by extending network lifespan and lowering routing overhead, critical for reliable packet delivery. This paper details the integration process of decision trees into the AODV protocol and highlights the substantial enhancements in network efficiency. The results affirm the effectiveness of this approach, advancing the capabilities of IoT network systems and enhancing overall network reliability.