Dynamic adaptive learning mobile agents for multi-modal input collaboration in home environment

Home server is developed to play the pivotal role of managing all home devices and multimedia content services through home networks. The possibility of a single user input device and single security agent to serve all the needs of the home is conventional to cope with the new technology revolution. The home server has to be enhanced to facilitate multi-modal inputs with multiple integrated user agents working together to facilitate all users at home. Processing of multiple input agents for dynamic context-aware output facilitates the system to learn adaptively using a probabilistic inference mechanism. We devise a hybrid collaboration architecture using fuzzy and neural agents to enhance our system operation to the trends of next generation home user requirements. Security in such a dynamic infrastructure is challenging. We devise mobile agents at each interface input device to manage multiple users using the mobile agent digital signatures to dynamically adapt to the context. The mobile agents are defined to operate the tasks on any mobile or stationary home device and collect the information needed for the home server. The system adapts to the dynamically changing user requirements and security context and produces the output based on the user´s context