Research On User Brhavior Analysis And Content Chaheing Strategy In Mobile Opportunistic Networks

Along with the rapid development of wireless communication technology and the widespread popularity of portable devices,the probability of which the use of smart devices in the same area has increased dramatically which leads to the explosive growth of network traffic demand.The above problems can be solved by building more cellular base stations.However,this method not only produces huge construction costs,but also requires a lot of manpower in the process of later maintenance.In addition,enlarging base stations can just temporarily solve the pressure of network traffic,and can not fundamentally solve the problem.The advent of mobile opportunistic networks is currently one of the most viable solutions to solve this problem of traffic explosion.By offloading data to the edge nodes it can not only reduce the construction of the base station,but also effectively solve the cellular networks congestion problem.Since the development of mobile opportunistic networks is still in its infancy,there are still a lot of problems especially in the user behavior analysis and buffer management,such as the distribution of contents in data unloading process,the placement of contents when the buffer is limited,the influence of user social relations on content forwarding,and so on.In view of the above problems,this paper studies the user behavior analysis in the mobile opportunistic network and the content caching management mechanism of the node.The main contributions are as following:(1)Analyzing the user contact situation from the perspective of social relations.Under the mobile opportunistic network environment,it divided user into two social relationships: friends and strangers.According to the number of contacts between users and access points,we classify the user area,and by analyzing the contact differences between two kinds of social relationships in different regions,we can make reasonable predictions for subsequent user contact behaviors.(2)Content caching management.We proposed a combinational buffer management strategy with message migration method based on neighborhood cooperative caching.In addition,the strategy fully considers the related characteristics of the message and the node when calculating the utility value,thus a new sort of message sorting method for the node buffer is proposed.The results of simulation experiments show that this strategy can improve the delivery rate of network and reduce the average transmission delay of data.(3)In view of the pressure caused by the data surge to the backbone network,we proposed a content buffer management strategy based on data offload.This strategy obtains the number of message backup by maximizing the hit rate of the message when the node buffer is limited,and calculates the message delay gain by combining the contact probability between the nodes.Content scheduling strategy based on 0-1 knapsack algorithm,we can dynamically solve the problem of message allocation in the network,thereby improving the overall performance of the network.This dissertation focuses on three aspects of mobile opportunistic network: user behavior analysis,content caching management strategy and data offloading.We prove the effectiveness of the research content by presenting corresponding strategies and network models,as well as analysis of relevant theories and simulation experiments.