Joint peer discovery and resource allocation for social-aware D2D communications: A matching approach

With the unprecedented growth of the mobile data traffic, device-to-device (D2D) communication has emerged as a promising solution to relieve the heavy burden of mobile terminals on the traditional cellular networks. However, how to jointly optimize the allocation of users, contents, and spectrum resources remains uncertain. In this paper, we address the joint peer discovery and resource allocation problems by combining both the social and physical layer information. The social relationship, which is modeled as the probability of selecting similar contents and estimated by using the Bayesian nonparametric models, is used as a weight to characterize the impact of social features on D2D pair formation and content sharing. Next, we propose a three-dimensional iterative matching algorithm to maximize the sum rate of D2D pairs weighted by the intensity of social relationships while guaranteeing the quality of service (QoS) requirements of both cellular and D2D links simultaneously. Simulation results show that the proposed algorithm is able to achieve more than 90% of the optimum performance with a computation complexity one thousand times lower than the exhaustive matching algorithm. Simulation results also demonstrate that the satisfaction performance of D2D receivers can be increased significantly by incorporating social relationships into the resource allocation design.