Research On The Smart Home System Based On Context-aware Computing

So far, remarkable achievements have been made since the concept of Smart Home came into being in the 1980s, such as the X-10 System from the USA, EIB System from the Europe, 8X system from Singapore, HBS from Japan and the e-Home from Tsinghua China ect. However, most of the researchers focusing on the study of the network technology instead of building human-centered & service-oriented living environment failed to reveal the essence of services in the Home System. With the expansion of the global informationization tide and the rise of the household network, people's home-life is gradually developing. On the other hand, the rapid development of network technology and the increasing interdiscipline cross-applications of artificial intelligent technology also provide ways of new thinking for Smart Home System's design & realization. According to the "human-centered" essential character of pervasive computing, constructing the Smart Home System based on context-aware computing is propitious to solve the open and collaborative problem about Home System in existence. Aiming at the goals above, this research focused on several key issues as follows:①Research in depth on the two natures of pervasive computing——providing transparent services to people and context-aware computing, discussed detailed context, context-aware computing, the status quo and related applications.②Construction of context-aware computing Smart Home System demands highly for integration, intelligence, openness and autonomy. Then, considering the advantage of Agent, we put forward to construct of the Home System by using the Multi-Agent technology, and discussing the function, communication and workflow of each Agent in it in detail. With JADE, we have accomplished the design of the systematic experimental model, and probed into the instantiation of this system.③From the target of learning users'habits, we use Hidden Markov Model to model the context of domestic environment and make reasoning, revealing the users' habits. As there is strong correlation among a single user's activities, the article has extended the traditional model and the corresponding algorithms, through experiments validated that the extended Hidden Markov Model superior to the traditional model. Combining context-aware computing with Multi-Agent technology will apply the research to Smart Home System, which is not only the realistic need, but the inevitable trend of theoretic development, to increasingly be paid much attention.