<![CDATA[Vehicular Ad-hoc Networks (VANETs) are dynamic wireless systems enabling vehicles to exchange data such as speed, location, and direction. A core component in VANETs is the routing protocol, with Ad-hoc On-Demand Distance Vector (AODV) being widely used. However, AODV struggles in highly dynamic environments, leading to reduced performance in throughput and end-to-end delay. This study proposes an improved AODV protocol using the Learning Automata approach (LA-AODV), which enhances relay node selection for better routing efficiency. The performance of LA-AODV is evaluated against standard AODV using NS-3 and SUMO simulations, focusing on key Quality of Service (QoS) metrics: throughput, packet delivery ratio, packet loss, and end-to-end delay. Simulation results show that LA-AODV significantly increases throughput (up to 90.84 Kbps) and reduces end-to-end delay by 20–35%, while lowering communication complexity (15–25% fewer Flow IDs). However, LA-AODV suffers from higher packet loss (up to 62%) and lower delivery ratios (37–92%) compared to AODV. This performance trade-off highlights LA-AODV's improved data efficiency at the cost of reliability. To mitigate delivery issues, solutions such as buffer tuning, adaptive queuing, or hybrid protocols are suggested. Future research should address these challenges and examine LA-AODV's scalability in larger VANET deployments.]]>
Copyrights © 2025
