Social Network Information Based Radio Resource Management Of D2D Communications System

The combination of Device to Device (D2D) and social network is one of the hot topics in the current and future communication industry. With characteristics including wide coverage, low cost, high system capacity and spectral efficiency, the D2D technology is one of the key technologies of the communications network in the future.However, nowadays the D2D communication is mostly designed and optimized based on the traditional physical environment, and the influence from the unique characteristics of the social network is ignored which is constructed by the device holders. For example,the social relationships between users in the social network not only reflect the normal probability D2D device can communicate properly, but also affect the transmission power of the repeater which assists D2D communication. To further improve the communication performance of D2D communication, we need to optimize the current D2D communication system combining with social networks.Three major research directions are studied in this thesis, including research on social network community detection algorithms with high accuracy, community-based D2D system spectrum allocation and D2D relay selection based on social networks. The research contents and innovation points are summarized as follows.1. In the study of the social network community detection algorithm with high accuracy, with respect to social relationships among different users in the strength of the different social networks, we proposed a community detection algorithm based on the weight of network characteristics. By using this algorithm, the social relations are maped through multiple paths, to convert the original social networks into a non-weighted network, so as to achieve high accuracy of community detection. Combined with the literature data and the actual statistics of deep processing, the proposed algorithm was tested and the results show that it is more accurate than the traditional methods.2. In the research of community-based D2D system spectrum allocation, based on communications and the social double layer system model, a strong relationship between social users in each community is ensured by distributing the D2D users and cellular subscribes into different communities according to social relationships, D2D users are able to use spectrum resources of cellular users effectively. The application of resource allocation based on social networks in the community improved system performance and reduced the complexity of the algorithm. From the simulation results, the proposed spectrum resource allocation algorithm can effectively improve the throughput capacity of the system.3. In the study of D2D relay selection based on social networks,we proposed to combine online social network content distribution and channel gain of communication layers to give a high reliability and low latency of relay candidate set. Then based on the social relations between the relay users and source users, the proportional relationship between the strength of the relationship and the power of the relay device is obtained. Based on the aboved two steps, we proposed a relay selection scheme to optimize the outage probability. Simulation results validate that the proposed relay device selection scheme can effectively reduce outage probability of D2D communication and improve the throughput capacity of the system, and further evident the scheme to be more suitable for dense cellular networks, and can be applied to dense complex communications networks in the future.