Research On Node Mobile Model Of Opportunity Social Network

With the continuous development of portable intelligent devices, new requirements forthe development of network technology are continuously put forward by society. Beingdriven by such new demands, opportunistic social network as an emerging field is beingpaid more and more attention from the researchers. Opportunistic social network refers tonetwork built with mobile equipments carried by people with social relationship, data canbe exchanged when the terminal equipments carried by people move around. Fixed networkinfrastructures are not necessities in the opportunistic social network, so it is very flexibleand can be built in a short time. The network has been widely applied to various occasionssuch as social relationship mining, city planning. Moreover, the network technology witha promising future can be widely used in information perception and transmission, etc.Because it is so hard to build the real opportunistic social network to evaluate thenetwork performance in the real scenario, the simulation research is the main methodadopted in the research for opportunistic social network. During network simulation, themovement mode and pattern of nodes can be provided by mobile network which is one ofthe most important components in the simulation. The main characteristics of socialnetwork is that mobile communication equipments are carried by people in the network, sonodes in the network can move the way people do. In this paper, two different nodesmobility models are presented based on the analysis and summary for human movementcharacteristics, combining with the related opportunistic network theory.1) A kind of interest community guided nodes mobility model is presented aiming atthe social movement characteristics showed by human in some social service activities,combining with the theory of social network node centricity. The degree of people’s interestin some events are represented by the nodes’ interest probability value vector, at the sametime the changes in nodes’ interest probability value are described with interest model of human dynamics, thus the nodes are of wide interests and time-varying. Firstly, the nodes inthe simulation scenario are put into different communities, the super node, which isresponsible for collecting information in the community, is the one with the highest valuecalculated according to the node degree centrality. Nodes in the community follow therandom waypoint mobile model (RWP), in which roaming nodes move to the communityunder the drive of interest, stay within the target community for a while and move to nexttarget community.2) A kind of nodes mobility model based on people’s social activities is presented,aiming at the driving effect of people’s social relationships and interest on nodes’ socialactivities in real life. Social activities only refers to people’s social activities in free time.For these activities, people tend to choose someone with whom they have good relationshipand with the same interest and move together, and tend to go a nearby place where theyalways go to. According to the mobile way in such a application scenario, in the model threekinds of relations are established: nodes and nodes, nodes and activities and activities andactivity places. Firstly, nodes’ relations set is obtained by the input of social network model.Then the degree of nodes’ interest on the activities are described with the interest probabilityvalue which changes periodically and randomly over time. Pearson correlation coefficient isused to calculate the node and node interest similar set, according to the interest set similarto what will happen at the choice of the final choice activity places.While the above two models for different scenarios of social activities, but reflect thedriving effect of people’s social relationships and interests on mobile nodes. At the sametime, simulation results show that the two models are consistent with the statistical resultsof real data sets.