Resource Allocation for UAV-Assisted MIMO-NOMA Wireless Caching Networks

With the advantages of high flexibility and low cost, unmanned aerial vehicles (UAV)-assisted communications have been recognized as an important technique in the future wireless networks. In this paper, we investigate resource allocation strategies in a UAV-assisted multiple input multiple output non-orthogonal multiple access (MIMO-NOMA) wireless caching networks (WCN) to minimize the user delay. Since the optimization problem is a mixed-integer non-linear problem, it is decomposed into three sub-problems, which are UAV deployment, hybrid beamforming and power allocation. For UAV deployment, we apply K-means method to group users according to users' distribution to determine the two-dimensional position of the UAVs. Since the base station (BS) to UAVs are millimeter wave (mmWave) MIMO channels, DFT codebook and zero forcing (ZF) are used for analog beamforming and digital beamforming, respectively. In the content delivery phase, we optimize the power allocation factors for NOMA users using the genetic algorithm (GA) to minimize user delay. Simulation results confirm that the proposed NOMA scheme can achieve better performance than traditional frequency division multiple access (FDMA) scheme.