| Opportunistic networks is a kind of new network model which rely on the nodes' mobility bringing the encounter chances.Communities structure of network is basic to study network structure,reflecting the network's function and features.It has a significant impact on analyzing the networks' features.However,in opportunistic networks,the nodes are moving at every moment,its communities aren't the same as static communities in real networks.So detecting the dynamic communities in opportunistic is an import issue.In order to solve the issue that detecting the dynamic in opportunistic networks,after analyzing the current detecting communities algorithm in opportunistic networks,proposing an algorithm called Memory-based cognitive heuristics.Node remembering the historical contact information with other nodes,calculating the memory activation with other nodes and dealing with them,putting the nodes which are in same memory activation threshold range to the same community,when every nodes has a community ascription,then completing the community detection.For the sake of saving the buffer memory space of the nodes,a forgetting process was used in this paper.For the historical contact information with other nodes,the earlier contact information is,the less impact on the memory activation,it should be forgotten by the node.If some historical contact information reaches the forgetting condition,then the node delete this contact information from the buffer memory space.In order to verify the significance of the detecting communities in opportunistic networks and used community better,an incentive strategy was proposed which is based on community and reputation.In order to verify the algorithm performance,a simulator called ONE was used for experiment and simulating.Comparing the performance with the classical algorithm proved that this algorithm could detect the communities in opportunistic networks relatively accurate in the opportunistic networks. |
PDF Full Text Request