Self-organized Resource Allocation Based on Traffic Prediction for Load Imbalance in HetNets with NOMA

Abstract

With the development of mobile communication technology, the data traffic of wireless cellular network has grown rapidly in the past decade. Because of the various bandwidth-eager applications and users movement, load imbalance has become an increasing severe problem, impacting the user experience and communication efficiency. Especially, it may lead to the degrading of resource utilization and network performance. In this paper, we investigate this problem and propose a self-organized resource allocation algorithm that allocates the resource to somewhere that the resource is needed to deal with the load imbalance problem. The typical heterogeneous network with non-orthogonal multiple access (NOMA) is discussed. A traffic prediction model is applied to the NOMA system. Then the self-organized resource allocation is formulated as a mixed integer non-linear programming (MINP) problem aiming at maximizing the overall throughput. The optimization problem is hard to tackle so we propose an algorithm to obtain a suboptimal solution via quantum-behaved particle swarm optimization (QPSO) algorithm. To evaluate how the resource is allocated according to the data traffic requirements, an indicator called evolved balance factor (EBF) is proposed to jointly consider the resource utility and the distribution of data traffic. Simulation results show that the proposed algorithm achieves a better performance in the overall throughput compared with exiting schemes.