Socially-inspired Peer Discovery And Resource Allocation For Device-to-Device Communication

With the popularity of smart terminals and the explosive growth of mobile services,it has placed a heavy burden on traditional cellular networks,such as shortage of spectrum resources,overload of base stations,and congestion of communication links.Thus,D2D(Device-to-Device)technology came into being.As important technologies for D2 D communication systems,peer device discovery and resource allocation have received great attention.However,the design and optimization of existing D2 D communication are mostly based on various conditions of the physical domain,and do not consider the influence of the social domain.Generally speaking,communication devices are carried by users,and users spontaneously form relatively stable social relationships(friends,colleagues,etc.)through social behaviors.To some extent,these social relationships may affect people's communication behavior.Therefore,in order to further improve the performance of D2 D communication systems,combining social networks with D2 D communication has become a new research hotspot.In this thesis,the following three aspects are investigated,including the analysis and quantification of social attributes,device discovery based on social attributes and resource allocation based on social attributes.The main research contents and innovations are as follows :1.In the analysis of social networks,the social attributes are divided into local attributes and global attributes.Local attributes include communication intervals,communication durations,and social content,and global attributes include centrality and community.Moreover,pajek software is adopted to visually analyze real social data sets to obtain visual graphs of social networks,and further analyze the beneficial effects of social attributes in social networks on peer device discovery and resource allocation.2.In the research of peer discovery based on social attributes,in order to solve the problem of link insecurity and ignoring the user's personal will in the existing solution,we proposes a novel peer and service discovery scheme.Three key social attributes are identified and quantified according to the historical record and the social attribute,including the trust degree,the similarity degree between users and the center degree of each user.Then,based on above three criteria,function utility is proposed to unify the above three criteria,and compare the results with the predefined threshold and put the larger into the neighbor list of the target user.Finally,the target user performs the peer discovery among the neighbor list.Simulation results show that,our scheme significantly improves the peer discovery ratio and link security.3.In the research of resource allocation based on social attributes,considering that the existing scheme does not consider the complex social relationship between users,and can not maximize the overall utility of the communication system and improve the throughput of the system,a novel resource allocation scheme is proposed.We divides D2 D users into different user groups by leveraging the user's centrality and social relationship.Centrality is the criterion for selecting different groups of core users,other users are divided into different groups according to the strength of social relationship with core users,and then each D2 D group is regarded as one to multiplex the spectrum resources of users in the cellular network.Assume that each cellular user can only share resources with one D2 D group,calculate the sum of the transmission rates of D2 D users and cellular users under each multiplexing scheme,and allocate resources according to the principle of “maximum transmission rate”,which not only ensures the communication quality of users,but also improves the throughput of the system.Finally,the simulation experiment is implemented,and the results show that the proposed resource allocation scheme can effectively improve the system throughput.