System Modeling And Prediction Research Of Context Awarenwss In Smart Space

As the typical application of ubiquitous computing and the developing tendency of mobile commerce, Smart Space has become the frontiers top of academic circles at home and abroad. Domestic and foreign research institutes have launched lots of researches on Smart Space and Context Awareness, both theoretically and practically. There are still few researches on Open Smart Space (OSS) and its system architecture, which is the second stage of Smart Space and emphasizes the mobility of mobile devices greatly. What's more, the weak grasp of key points and the simplicity of prediction mechanism in OSS, like mono-dimension mechanism based on position or bi-dimension mechanism based on position and service, can just not meet with people's increasing needs.With the study of previous researches on Smart Space and Context Awareness, this paper designs a Context Awareness system architecture in OSS, which consist of Contexts subsystem, Users subsystem, Services subsystem and Prediction inference engine. And with considering the features in OSS, this paper presents and analyzes the three key attributes in Context Awareness-position, time and service.Subsequently, with analyzing the integrated log files of location, time and service of users in context-aware environment, this paper improves the N-gram model, advances two prediction algorithms (N-gram-A and N-gram-B), and compares the advantages, disadvantages and applicable scope of these two algorithms. In the process of using these two algorithms, we can alter the prediction rules, the value of n in N-gram and the segments of time dimension to match needs of users with the current resource in an optimized way. In order to help mobile users obtain more satisfying services and the operators get more profits, the article aims to use the context-aware data effectively and we hope it can play a positive role in both academic research and business practice.Finally, based on the preceding analysis and research, this paper makes some proposals in developing our national Smart Space, respectively from the government policy level, the industrial development level, the academic research level.