NOMA Enabled Resource Allocation for Vehicle Platoon-Based Vehicular Networks

Abstract

This paper considers a traffic scenario with hybrid driving platoons and single driving vehicles on automatic driving age, and investigates Non-Orthogonal Multiple Access (NOMA) enabled resource allocation in platoon based vehicular networks for vehicle to multi-vehicle (V2mV) communication. The resource allocation problem is formulated as an optimization problem with multi-constraint. To solve the mixed-integer nonlinear programming (MINLP) problem, we decouple this problem into power allocation and sub-channel allocation. For power allocation, we propose a lane condition based power allocation scheme, which can provide more equitable capacity for platoons among different lanes. For the latter, a multi-agent Q-learning framework is proposed and which can achieve better convergence performance. Finally, simulation results verify the effectiveness of proposed algorithm for NOMA enabled resource allocation in vehicular networks.