Resource Allocation Based on Three-Sided Matching Theory in Cognitive Vehicular Networks

In this paper, we investigate the resource allocation and vehicle to everything (V2X) offloading in the cognitive vehicular networks. The cognitive radio (CR), mobile edge computing (MEC), and non-orthogonal multiple access (NOMA) schemes are applied aim to solve the combinational problem of resource allocation and V2X offloading. The problem for jointly optimizing power and time allocation in the MEC based CR (CR-MEC) networks is conceived. We decompose the joint optimization problem into two subproblems, which are power allocation and time allocation problems. In order to solve this joint optimization problem, an advanced comprehensive resource allocation (ACRA) algorithm based on three-sided matching theory is employed. More specifically, the proposed algorithm is to realize the most reasonable matching among primary users (PUs), cognitive users (CUs) as well as a cognitive base station (BS), and put forward a V2X offloading strategy, by appropriately allocating power and time aim to minimize the system energy consumption. The simulation results show that, our proposed algorithm converges to stable. Furthermore, the proposed NOMA based CR-MEC networks can achieve lower energy consumption compared to the orthogonal multiple access (OMA) based CR-MEC networks.