The Research Of Big Data Chunk Disseminating In Mobile Opportunistic Networks

A Mobile Opportunistic Network(MON)is an intermittently connected network which lacks an online connected route between the source node and the destination node.Due to the mobility of nodes in the network,the connection is unstable and interrupted,and the node is also of limited capability in its energy consumption,communication capacity and storage capacity.These features make data transmission extremely challenging in mobile opportunity networks.While the existing routing methods pay less attention to the actual size of the data going to be transmitted,the storage space of the relay node and the communication bandwidth between the nodes must be considered during the transmission of big data chunks.At the same time,the security of the data also needs to be taken into consideration.In the face of big data chunk routing in mobile opportunistic networks,this thesis selects several representative scenarios to do research on.Firstly,the author proposes the multi-copy routing method with the limitation of storage capacity and communication bandwidth for the transmission problem of big data chunk in MONs.Secondly,a multicast strategy based on the distribution of data packets on the auxiliary equipments is proposed.Finally,a security and efficient broadcast method for big data chunk broadcasting in a mobile social opportunistic network is proposed against malicious users' attack.The main contribution of this thesis is three-fold:(1)Taking a VANET(Vehicular Ad-hoc Network)as the specific research object,the single-copy data forwarding between vehicles is firstly modeled as a Multiple Knapsacks Problem and then extended to the multi-copy scenario.The simulation experiments on the actual data set are performed where the data packet is large and the communication capability between nodes and the storage capacity of the nodes are limited.(2)The distribution strategy of big data chunk on the Throwboxes is studied in the network.In the case of limited Throwbox memory,the problem of allocation of large data packets on Throwbox is modeled as a Bipartite Matching,which maximizes the success rate of delivery.The simulation experiments performed on real data trace proved that the proposed method works well.(3)Aiming at the attack behavior of malicious nodes in the mobile social opportunistic network,SEBMo So(Secure and Efficient Broadcasting in Mobile Social Networks)is proposed to ensure data transmitting security and efficiency by forwarding data using malicious nodes and secret-sharing schemes.Simulation experiments are performed using real data trace.