Connection manager using FAHP of MCDM techniques in heterogeneous network

Connection Manager Intelligence Agent´s (CMIA) an attractive feature is use the network traffic and behavior with multi-attributes to find the best network devices and connection devices meet user habits that it possible to access various network contents everywhere at any time. In this agent, information process is thoroughly integrated into embedded system for connection manager of MeeGo or Android. This paper proposes a new and generic algorithm for connection manager in the heterogeneous wired or wireless environments. The proposed algorithm uses a Fuzzy Analytic Hierarchy Process theory (FAHP) of Multi-Criteria Decision Making (MCDM) system to achieve scalable, adaptable, and flexible solution. This study implemented an interactive interface through of this function acquisition user behavior and to scholars of the machine used a boosting iterations for User-Case, the algorithm maintains a set of weights as a distribution classes table of case as in the parameter learning by user-case and its quite possible that the Expectation-Maximization (EM) of Maximum Probability Model (MPM) by classification from user behavior. User interaction showed that the agent satisfactorily matched user intent.