Exploring spatial motifs for device-to-device network analysis (DNA) in 5G networks

Abstract

Device-to-device (D2D) communication is a promising approach to efficiently disseminate critical or viral information across 5G cellular networks. Reaping the benefits of D2D-enabled networks is contingent upon choosing the optimal dissemination policy and resource allocation strategy subject to resource, topology, and user distribution constraints. In this paper, a resource allocation problem within a single-cell orthogonal frequency division multiple-access (OFDMA) network has been formulated to optimize the system throughput. The problem is cast as a mixed binary integer programming problem, which is challenging to solve. Therefore, in order to obtain a sub-optimal solution, the optimization problem is decomposed into two subproblems: seed selection and subchannel allocation. For the seed selection sub-problem, a novel D2D network analysis (DNA) framework is proposed to explore frequent communication patterns across D2D users, known as spatial motifs, to determine the optimal number of devices which can disseminate contents. Furthermore, to solve the second sub-problem, a heuristic algorithm is introduced. Simulation results show the effectiveness of exploring motifs to design D2D dissemination policies, and show that the proposed algorithm can achieve a performance gain of up to 40% compared with a baseline scheme that selects subchannels randomly.