Research Connectivity Factors Model In Opportunistic Sensor Networks

Opportunistic Sensor Networks(OSNs) is self-organization network that doesn’t need a complete path between source node and destination node. The mobile nodes opportunistic meeting achieve to network communication in the OSNs. OSNs have the characteristics of the early delay tolerant networks and intermittent connected networks, which have obvious message delay in the process of forwarding, and high error rate during the transmission. It’s suitable for delay-insensitive, infrastructure-less, quickly setting up network applications, such as wildlife track, vehicle network, urban and rural network, etc. There are many reasons that make networks become sparse distribution, such as physical constraints, cost considerations and node failure owing to destructive events. OSNs may become isolated when achieve the large-scale sensing. The networks connectivity is achieved by mobile nodes. It is a challenge how to research the OSNs connectivity in this scenario.Currently, the OSNs connectivity has variety definition. Usually, the network connectivity metrics can reflect the network connectivity. In the thesis, the state of art of MANETs connectivity and structure of OSNs are introduced. According to the features of the multi-region OSNs, the OSNs layered model is designed. On the basis of analyzing the existing network connectivity metrics and connectivity parameters’ research, the network message delivery success rate and average delivery delay are defined. They act as the layered model of OSNs connectivity metrics. The distribution of network connectivity metric mean and variance in different regions and the relationship between network connectivity metrics and connectivity parameters are analyzed, on the basis of many experiments.Considering connectivity parameters have relevant relations, the SPSS is utilized to analyze the connectivity parameters’ correlation. The factor analysis method is used to extract the main parameters as fitting factors.On the basis of above research, the wavelet neural network based on particle swarm optimization algorithm is utilized to fit the connectivity parameters, and get a OSNs connectivity parameters model in the thesis. The simulation results show that the connectivity parameters can be available in the different mobile models and the connectivity parameters model is valid. In order to further verify the validity of the model, the Telos B nodes and intelligent cars are utilized to set the scenarios of the multi-region OSNs. The experiment results show that the model can effectively reflectthe multi-region OSNs connectivity.