Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Software Engineering > Software Development

Research and Application on Gradation Modeling for Adaptive Middleware Entity

Author JinLingMin
Tutor WanJian;WuQing
School Hangzhou University of Electronic Science and Technology
Course Applied Computer Technology
Keywords Pervasive Computing Reflective Semantic Component Extended Semantics Virtual Agent Ontology
CLC TP311.52
Type Master's thesis
Year 2009
Downloads 29
Quotes 0
Download Dissertation

Development of information technology integrates intelligentizing, miniaturizing, and human-computer interaction of computing devices, resulting in a new "human-centered" computer model - Pervasive Computing. In the ubiquitous computing environment, people are able to acquire services“anytime”,“anywhere”and "transparently", which would require the basic software to provide interactive mode that is independent of time and space. A new middleware with adaptive features is needed to replace the traditional middleware in that traditional middleware, as the first floor software of Pervasive Computing, cannot meet dynamic state, heterogeneous, open and intelligent features as expectations in respect of software architecture, theoretical models and methods. At present, how to set up an adaptive middleware model to adapt to the pervasive computing environment and propose the corresponding supporting methods and theories have become hotspots and difficulties in academic research in the field of pervasive computing.There are two pressing questions in the study of adaptive middleware model for ubiquitous computing. First, along with its development, component technology has been closely integrated with the middleware technology, and widely used to the design and development of middleware software. How to build an adaptive middleware model which is centered on core entities of semantic components with introspective characteristics in this case? The second is semantic components can be combined into a larger particle of composite components, which have richer function, more complex internal logic in modeling adaptive middleware. How to describe composite components and model self - adaptation middleware from the aspect of composite components?This article revolves the study of problems in self-adaptive middleware modeling, which includes:Firstly, we study the problem of component’s semantic information description. On the point that the traditional component model is lack of rich semantic information description, so we use the ontology technology to formally describe the component semantic information to form the semantic component model. Semantic component can be understood by the system, thus can be derived and assembled automatically. But there are still "Black-Box", "Static" and other problems existing in the semantic components, which could result that semantic components can not be partly dynamically adjusted to adapt to changes in demand. We apply the reflective technology into Semantic Component to make the Semantic Component can partly dispark its internal realization, thus turn into the“Grey-Box”featured Semantic Component. Based on this, we proposed a Semantic Component Model based on Reflection. Reflective semantic components with introspection features which have insight into the information of their own internal status and adjust according to the specific circumstances to adapt to the changing running environment.Secondly, based on the reflective semantic component model, we introduce the Agent technology to build the Extended Semantics Virtual Agent. With Agent-BDI theoretical structure model, we descripted ESVA in formal. Then we research the interactive problem between ESVA in the ESVA Union. To solve this issue, we define the Semantic Interactive Protocols (SIP) for the interaction between ESVA, thus can descript the interaction between ESVA in formal. SIP includes join, depart, cooperate protocol.Finally, we examine the validation of the Reflective Component Model and Extended Semantics Virtual Agent through adaptive middleware in the mobile video system. Keywords: Pervasive Computing,Reflective Semantic Component,Extended Semantics Virtual

Related Dissertations
More Dissertations