The Key Technology Research On Data Dissemination Mechanism In Hierarchical Opportunistic Network

Opportunistic network is a kind of wireless self-organized networks in which the communication between source nodes and destination nodes is achieved under partially connected network environment. The fact that the application scenarios of opportunistic transmission become more and more widely, brings new requirements on network architecture, data dissemination mechanism and mobile model.To improve delivery ratio, reduce network delay and power consumption, aiming at designing a novel architecture and data dissemination mechanism in opportunistic networks, we investigate the data dissemination mechanism in hierarchical opportunistic networks by means of Optimization Theory, Markov Chain and Game Theory.This work is supported by the National Natural Science Foundation of China 《Research on the hierarchical opportunistic network based on the distributed storage》 and other national projects. We further study the data dissemination strategies in the hierarchical opportunistic network architecture and explore the solution of new scenarios. The major contributions of the paper are listed as follows:(1) Current mobility model cannot highlight the behavior characteristics of mobile nodes, this paper proposes a time-variant opportunistic network mobile model, which divides different sub-movement models according to the different time periods. And compared with the actual data traces, it is proved that the time-variant mobility model fits into the node behavior characteristics in real scenarios by analyzing some parameters, i.e., residence time and transition frequency of the destination node. For the scalability problem of opportunistic network, we propose an adaptive cluster head selection mechanism based on hierarchical opportunistic network architecture, providing the architectural foundation for the study of the rest chapters. Simulation results show that the proposed scheme reduces energy consumption and prolong the network lifetime.(2) Based on two kinds of heterogeneous network model, this paper shows the inter-contact time between nodes not entirely obey exponential distribution and compares the message forwarding delay performance of direct-forwarding and three-hop routing protocol in the individual heterogeneous network model, spatial heterogeneity network model and the homogeneous network model. It is proved that there exists forwarding performance difference between the two heterogeneous network models. Based on the study of node mobility, we propose a transmission probability matrix by discrete Markov process, and establish the connections between the social structure and the individual nodes to design the data dissemination strategy based on the next hop node selection and the entire network information, which improves the message delivery ratio and reduces the forwarding delay and network overhead. Meanwhile, combined with community detection, system model and mobility traces, considering the maximum number of message copies, we design a data dissemination mechanism based on forwarding and replication algorithms in the opportunistic network. This mechanism improves the message delivery ratio and reduces delay under the conditions of the forwarding overhead restrictions.(3)For Epidemic opportunistic network, in order to prolong the network survival time, the paper proposes Epidemic routing model solutions, Epidemic routing defect solutions and the constrained-Epidemic-based control algorithm of opportunistic network copy. And it also deduces the Game function curve on energy and delay and shows how the game function curve applied to the copy control mechanism of constrained Epidemic opportunistic network. Based on the residual energy information and restrictions on the number of copies, this paper proposes a new copy number control energy balanced scheme in constrained Epidemic opportunistic network. The simulation results show that the scheme can effectively prolong the network survival time.(4)In order to reduce the data traffic load of infrastructure, we investigate a novel infrastructure-based data dissemination scheme in opportunistic network. The basic idea is to stimulate the users with delay-tolerant and offloading capabilities to participate in the infrastructure-based traffic offloading, and offload the traffic onto opportunistic network. In order to describe the dynamic characteristics of delay tolerance, we design an incentive mechanism based on reverse auction. In addition, in order to predict the offloaded traffic through opportunistic network, a prediction mechanism is proposed. Simulation results show the validity of this mechanism for traffic offloading through opportunistic network and it provides an effective novel method for future infrastructure-based network traffic offloading.In summary, this paper conducts an in-depth research on the hierarchical data dissemination mechanism of opportunistic network.