Research On Routing Algorithm Based On Node Social Identity And Social Influence In Opportunistic Mobile Social Network

With the rapid development of mobile devices and wireless communication technologies,the people-oriented Opportunistic Mobile Social Networks(OMSNs)have been beginning to rise,which do not need to rely on communication base stations for message transmission.OMSNs have the same series of characteristic attributes of Opportunistic Networks,such as frequent contacts among mobile nodes,complex and changeable network topology and limited network buffer resources.However,OMSNs also have some differences from the Opportunistic Networks,for example,the forwarding behaviors of mobile nodes are guided by the will of human beings.The movements and contact behaviors of mobile nodes obviously reflect the social attributes and relationships of people.Therefore,how to accurately analyze and make reasonable use of the social properties of mobile nodes is the most important task for researching opportunistic mobile social networks.Due to the limitation of the physical environments,the connection status of mobile nodes in OMSNs changes frequently.It is difficult to maintain a complete and stable communication link between source nodes and destination nodes,which makes the traditional communication mode unable to work under the network environments.To realize effective communications,it is helpful to use the "carry-store-forward" mode.In addition,how to select the appropriate relay nodes for message transmission is also the key to achieve the optimal routing performance.Most mobile nodes in OMSNs are composed of people who hold mobile devices with Bluetooth or Wi-Fi,and the social behaviors of people are controlled mainly by their stable social characteristics.Therefore,it can effectively improve the performance of routing algorithms by introducing the social characteristics.At present,researchers have introduced a series of social characteristics into the design of routing algorithms.But,it can be found that social identity,an indispensable social characteristic in human life,still lacks attention.Moreover,social influences that determine a person whether is able to influence and change others'behaviors are also not effectively used in the process of routing decision-making.To further improve the routing performance,we propose a routing algorithm based on the social identities and social influences of mobile nodes.The main contributions of this work are as follows:(1)Social identity extraction algorithm for mobile nodes in opportunistic mobile social networks.First,we define that the mobile nodes in OMSNs have multiple social identities.Early studies of social psychology theory show that when people contact and interact with others who have the similar social identities,they will acquire strong self-worth and social belonging.Similar social identities of people can give them a sense of self-identity and can also provide them with an effective social framework.Thus,we believe that social identities of people can further affect their social behaviors.Then,we propose an algorithm that extracts the social identities of mobile nodes based on the content attributes of their generated messages,and the membership degrees of their social identities can be calculated.Specifically,we define the mathematical model of the generated messages of mobile nodes in detail,and the definitions of the related concepts of message content attributes,message keywords,and message attributes correlation values are given.Furthermore,by using the related knowledge of fuzzy mathematics,we farther solve the fuzzy similarity matrix of generated messages of mobile nodes and the message attributes correlation values are calculated.Accordingly,the multiple social identities of mobile nodes and their membership degrees are obtained.Finally,the simulation results show that the proposed algorithm is correct and effective,which can extract multiple social identities of mobile nodes effectively.(2)A routing algorithm based on the social identities and social influences of mobile nodes in opportunistic mobile social networks.First,we introduce the concept of social identity into the opportunistic routing algorithms.Second,we describe in detail how to calculate the social influences of mobile nodes when they have multiple social identities.Specifically,the social influences of mobile nodes are related to their PageRank centrality degrees and influence factors.Furthermore,the influence factors of mobile nodes are determined by the encounter probability and trust degree with their neighbor nodes.Next,we calculate the forwarding utility values of mobile nodes based on their social identity similarity with the destination nodes and their social influence degrees.Then,the forwarding strategy of this routing algorithm is proposed:the messages will be forwarded to the relay nodes who have a higher forwarding utility value.Finally,the simulation results show that the proposed routing algorithm has good performance in terms of delivery probability,overhead ratio and average delay.