The Research And Implementation Of Application-Oriented Sensor Network Virtualization Technology

The Internet of things(IOT) is developing in the direction of diversification and openness. As the basis of the Internet of things, the sensor networks need to provide the infrastructure resources and flexible resource allocation for the diversified business. At present, Sensor networks are deployed in a specific sensing application with the heterogeneous characteristics of multi-vendor, that leads to the duplication of sensor infrastructure resource and the low utilization. In order to solve this problem, this thesis introduces the network virtualization technology into the sensor network, which is used to realize the sharing of sensing infrastructure resources.On the basis of the analysis of the existing problems, this thesis studies an architecture of application-oriented sensor network virtualization, building the virtual sensing network service layer above the original sensing infrastructure layer. Then, the thesis puts forward a corresponding solution of sensor network virtualization. To solve the communication problem between application server and smart sensor gateway and between smart sensor gateway and sensor nodes, this thesis introduces the Simple Network Management Protocol(SNMP). Combined with the limited characteristics of sensor network, Simple Network Management Protocol is lightweight adjusted, and the thesis expands the corresponding operations of SNMP.Finally, this thesis designs and realizes the solution of sensor network virtualization, including smart sensor gateway, sensor nodes, the classification scheduling mechanism and virtual abstract layer. And the solution solves the problems of resource scheduling and the co-existence of sensing applications. Finally, this thesis builds an experimental sensor network, and designs the different sensing applications. By analyzing and comparing the resource utilization in different application scenarios, as can be seen, the sensor network virtualization can realize the sharing of sensor infrastructure resource, and improve the resource utilization rate.