Resource Allocation In Socially Enabled Wireless Content Sharing Over D2D Links

With the development of wireless mobile communications, the contradiction between the increasing requirements for wireless services and the limited wireless resources is being aggravated. Recently, device-to-device (D2D) communications in cellular networks has attracted rising research interests due to its potential to solve the abovementioned problem.By enabling mobile devices to communicate directly without routing through the base station, D2D communications can improve overall system throughput, enhance spectral/energy efficiency, reduce content delivery delay, expand network coverage, etc.In this paper, the background and the significance of our work is firstly presented, and substantial literature research is done related to the techniques in heterogeneous cellular networks. Typically, this work investigates the wireless resource allocation problem in socially enabled wireless content sharing scenario in cellular D2D underlay. While satisfying the quality of service requirements of mobile users, this paper proposes a power optimization algorithm and a matching algorithm,achieving a good tradeoff between system performance and computational complexity.The wireless resource allocation problem is first investigated in cellular D2D underlay by considering only physical information and applying bi-partite graph matching theory. Under both full and statistical channel state information assumptions, the candidate set of users that can share the same spectrum is narrowed based on users' quality of service requirements. After that, the matching between D2D links and cellular spectrum is formulated as a bi-partite graph matching problem. A low-complexity algorithm is then proposed to optimize the transmit power and calculate the corresponding edge weight. After the construction of bi-partite graph, a low-complexity matching algorithm is adopted to allocate cellular spectrum to D2D communications to maximize the system throughput.In addition, a more practical scenario is considered in D2D-based wireless content sharing networks. In this case, the matching among the content helper, content requester, and cellular spectrum is investigated to maximize the system performance. Particularly, a heterogeneous bi-partite graph matching problem is formulated in this paper. By considering both physical conditions and social characteristics between mobile users, D2D links with high transmission success probability are selected for content sharing, and stable and reliable cellular spectrums are allocated for D2D communications. Theoretical analysis and numerical results demonstrate that our proposed scheme can reduce the computational complexity significantly with little performance loss.