Hypergraph Spectral Clustering Based Spectrum Resource Allocation for Dense NOMA-HetNet

In this letter, we propose a spectrum resource allocation algorithm in non-orthogonal multiple access heterogeneous network (NOMA-HetNet). We consider a dense two-tier HetNet consisting multiple macro base station (MBS) and multiple femto base stations (FBSs), which are very close to each other. All the MBS and FBSs are working in NOMA mode. Our hypergraph spectral clustering based method is proposed to deal with the serious cumulative interference. We aim at improving the system throughput with the prerequisite that all users’ fairness in the same base station is guaranteed. We analyze the computation complexity of the algorithm. The simulation result shows that our algorithm can improve the system throughput with low computational complexity.