Many-to-One Spectrum Reusing Resource Allocation for Device-to-Device Communications

Abstract: The deployment of cellular networks assisted by device-to-device (D2D) communications has great significance to improve spectral efficiency. In D2D communications underlay scenario, the D2D user equipments (DUEs) reuse the same frequency resources assigned to cellular user equipments (CUEs) for spectral utilization. Unfortunately, introducing D2D communications into cellular networks bring interference signals to CUEs and other DUEs. Therefore, to fully reap the benefits of D2D communications, efficient resource allocation with interference management is highly required. In this paper, we propose a system model for D2D communications underlaying cellular network, where multiple D2D pairs reuse the same frequency resources allocated to a single CUE. In addition, we propose a many-to-one frequency reusing resource allocation algorithm for D2D communications with the aim of maximizing the network sum-rate. To solve this NP-hard problem, we first propose an interference management algorithm based on graph theory to find the least interfered D2D pair groups, followed by proposing an optimized power allocation algorithm based on linear programming, and finally proposing a process matching algorithm to find the best sets of CUEs and D2D pair groups for spectral reusing. Simulation results demonstrate that the performance of the proposed many-to-one spectral reusing resource allocation algorithm exceeds the one-to-one resource allocation for D2D communications in terms of sum-rate.

Keywords: Device-to-Device communications, power allocation, interference management, matching process, graph theory, matching theory