An improved method of AODV routing protocol using reinforcement learning for ensuring QoS in 5G-based mobile ad-hoc networks

Abstract

5G-based MANET has received a lot of attention recently. Its fundamental feature is that nodes are constantly subjected to high traffic loads, while QoS requirements are extremely stringent. When applied to 5G-based MANETs, existing routing protocols have shown drawbacks. In this paper, we propose an enhanced AODV protocol solution for 5G-based MANETs. Using reinforcement learning, each node updates a state information database of intermediate nodes along routes to destinations. This database is used by the routing algorithm to find guaranteed QoS routes. Our solution is highly efficient in terms of throughput, end-to-end delay, and SNR, according to the simulation results.

Keywords

• 5G-based MANET
• Reinforcement learning
• QoS aware routing
• RLI-AODV