Based On Mobile Agent And The Optimal Search Theory Cbmr System

With the development of Internet and multimedia technology, more and more multimedia data, such as image, video, audio and 3D models, are presented through Internet. The keyword based search engines are not suitable for multimedia retrieval, which leads to the disorder of these data, and have deeply impacted on the application of multimedia. It is very important to develop content based multimedia retrieval systm.Because multimedia data processing is time-exhausting, it is difficult to do CBMR on Internet. Traditional C/S computing mode is not suitable any longer, a new distributed computing mode must be deployed, and mobile agent is such a computing mode.Mobile agent can solve the distributing problem in CBMR, but there still exist other problems. Given time is limited, the searching can not visit all multimedia data, so how to acquire the optimal searching result with the limited time is an interesting problem. We can use optimal search theory to solve this problem. Optimal search theory is a subject developed during the World Wide War II. The optimal search problem involves computing how resources can be allocated so as to maximize the probability of detection, or to minimize the comsumption of resources used. We use optimal search theory to optimize our CBMR system to acquire a higher searching result.We proposed a mobile agent based CBMR system model which is suitable for Internet, Intranet and P2P network. Our system model is a simplified one which is convenient for simulation experiment. Our research is the first one which use optimal search theory to optimize the resource retrieval system.We focused on the optimal mobile agent dispatching, the initial searching parameters dicision and updating mothods. We extended NS2 network simulator to support mobile agent simulation and did several experiments under this simulation...