Deep Reinforcement Learning Based Handover management for Vehicular Platoon

The 3rd Generation Partnership Project (3GPP) has maintained the standardization efforts on Vehicle-To-Everything (V2X) communications with NR-V2X (New-Radio-V2X) in Releases 16 and 17 in order to serve a wide variety of V2X use cases and applications with various quality of service needs and support situations with high vehicle density. One of the use cases that has attracted a lot of attention recently is vehicle platoons. Thus, vehicle platoons are a collection of vehicles that are tightly connected to one another virtually since they are going in the same direction at a constant speed while keeping a constant inter-vehicle distance. However, when a vehicle platoon travels from a source base station to a target base station, the existing communication technique incurs significant signaling overhead between the radio access network (RAN) and the core network (CN). In this study, we aim to reduce the number of handovers (HO) to enhance the performance of our vehicle network by using Deep Reinforcement Learning (DRL) as a technique. Simulation results reveal that the proposed model decreases the number of HO and increases the value of the cumulative reward.