Routing Algorithms Research Based On The Social Characteristics Of Nodes In Opportunistic Social Networks

With the deepening of the application and research of opportunistic social networks,it has important value for the Internet of Everything in the future 5G era.Based on the research of domestic and foreign researchers in the field of opportunistic social networks,this dissertation aims to achieve efficient data transmission based on the social characteristics of nodes in opportunistic social networks.Combining random graph theory,using the method of community division,and learning the inference epidemic model,on the basis of constructing the prediction of the local connected path,the influence of the true nature of the node on the network performance is further analyzed.The main research contents include the following two aspects:(1)Effective path-aware data transmission strategy in opportunistic social networks(EPADT),first of all,for the problem that the rapid movement of nodes in opportunistic social networks will lead to randomness and variability of the entire network,we analyze weight distribution among nodes and community reconstruction.Then,the connectivity of local network partitions in opportunistic social networks is studied in combination with the topological connectivity properties of the network.According to the connectivity of network partitions and random graph theory,the moving speed of adjacent nodes in network partitions and the average size of connected network partitions are calculated respectively.The stability and distribution law of locally connected network partitions can be known from the calculated velocity values and the average size of network partitions.Then,based on the above,two different situations are considered when selecting the relay node.The first is within a locally connected network partition,and the second is across multiple locally connected network partitions.In these two cases,the methods of extended ring search and calculation of interest similarity are used to determine who relays the data.At the same time,the service capability of the node is calculated.When the service capability is large enough,the message node will first copy the data and then transmit the data.(2)Malicious node detection and prevention strategy in opportunistic social networks(MNDP),firstly aiming at the social characteristics of nodes,taking opportunistic social networks as the research object,and socializing traditional information communication.The social characteristics of nodes are fully integrated into the selection process of relay nodes.Then,the network nodes mapped by social members are defined and a mathematical model is constructed.And classify the nodes in the network: general nodes,neutral nodes,malicious nodes and friendly nodes.Then,four types of node influence calculation methods are designed to reflect the influence ability of nodes.The friendly nodes among them can be selected as relay nodes by influencing the calculation value,and the malicious nature of malicious nodes can be detected at the same time.In addition,the EPADT and MNDP algorithms proposed in this dissertation use the opportunistic network simulation platform to conduct comparative experiments with the classic routing algorithms Epidemic,Spray and wait and Max Prop,as well as the relatively new ICMT,EIMST and FCNS in recent years.The experimental results are of great significance for the low time complexity of EPADT and MNDP algorithms,improving the success ratio of message delivery on opportunistic social networks,and effectively reducing network system delay and network overhead.