Resource Information Retrieval Method And Application Based On Name Service In The Internet Of Things

Internet of Things has gained rapid development in the recent years. Cisco's expert Dave Evans even preducts that in 2020, there will be 50 billion entity objects in the Internet of Things. This leads to an explosive increase of information contained in the Internet of things. It is very important to make reasonable and effective retrievals on information in the Internet of Things.At present, the retrieval methods in the Internet of Things are based on those used in the Internet, namely the key words based retrieval. But compared with the Internet, the Internet of Things has the following features: high information update frequency, large number of objects, big network size, multi-source heterogeneous data and complicated association relationships between data. If the Internet retrieval method is continued to be used, efficient retrievals of the resources in the Internet of Things would not be achieved.In order to solve this problem, this thesis carries out studies in the following three aspects:Firstly, due to the issue that the keyword based retrieval is not appropriate for the Internet of Things, an information retrieval method based on the name service is proposed. Name service is used to identify the entity objects in the Internet of Things, and is used to query the entity objects in the Internet of Things. Based on the name service, the resource retrieval system is established in this thesis. The system first establishes the rules to name the entity objects, then the resources are indexed based on the established naming conventions, and finally the retrievals on the resources can be done through the names in in the Internet of Things.Secondly, in the Internet of Things various data are often not well related. This raises the problem that multiple retrievals are required for getting the resource information of an entity object, and hence the retrieved data are often irrelevant. In this thesis, the proposed name service based retrieval method also takes the relation between the entity objects and the sensed data. Therefore, when the information of the entity objects are being retrieved, the corresponding sensed data are also provided. This helps to improve the retrieval efficiency.Thirdly, as how to realize the retrieval systems in practical application, this thesis takes the smart agriculture as the application scenario. The retrieval system has been applied in the smart farm of the Chinese Academy of Tropical Agricultural Sciences in the Huadu district, Guangzhou. With the help of China Intsernet of Things Name Service Platform, the proposed retrieval system has been tested in practice using the actual field data from the smart farm.Research shows that the proposed retrieval system based on the name service can significantly reduce the retrieval time when facing big data retrievals. The retrieval method considering the relation between the entity objects and the sensed data can improve the query efficiency. Being applied and tested in the smart farm scenario in real world also shows that the proposed retrieval system has value in practice.