Research On Keytechniques Ofthe Resource Addressing In The Internet Ofthings Based On DHT

With the popularity of computers, resource on the Internet has been far from meeting human needs. As a result, the development of Internet of Things has been a trend. The key condition to build it is to realize the function of resource addressing, which involves the entity coding standard and the resource addressing system. Nowadays, there are many organizations who do research on the Internet of Things. However, they hold on their own standards. What's worse, the existed resource addressing systems have limited abilities and lack the classical management to resource. The purpose of this research is to design a ubiquitous entity coding standard, and based on this standard and using improved Distributed Hash Table (DHT), develop a resource addressing system for the Internet of Things.Integrating the advantages of the entity coding standard named uCode designed by uID Center in Japan and the standard named Commerce Product Code (CPC) designed by the International Electronic Commerce Center of the Ministry of Commerce of China, a new coding standard is designed in this paper. This code has fixed length, belongs to the tree-level model, and distinguishes the type and amount of entities obviously. Based on the SkipNet designed by Microsoft, by improving the node's structure and the message routing algorithm, a resource addressing system is designed. The system is composed of two modules, named Entity's Code Addressing and Entity's Information Addressing. The former is responsible for translating the entity's code into the Uniform Resource Identifier (URI), while the latter is responsible for the joining, querying and exiting of the entity's code information.By the network simulation, the function and performance of the designed resource addressing system for the Internet of Things is tested in this paper. Experiment results show that this system has good functions on resource searching and node failure. With the amount of nodes increasing in the network, the average searching time and the average searching hops have a better O(logN) performance; The node failure mainly only affects the exited nodes and makes minimal effect on the non-failed nodes in the network.