An approach for designing cognitive engines in cognitive peer-to-peer networks

A cognitive network is a network which can learn to improve its performance while operating under its unknown and dynamic environment. Cognitive engine as part of a cognitive network tries to adaptively find an appropriate configuration for the network. Up until now no peer-to-peer network management algorithm has been designated utilizing cognitive networking concepts. In this paper, we adopt cognitive networking concepts and present a framework for cognitive peer-to-peer networks and then propose an approach based on cellular learning automata for designing cognitive engines for solving network management problems in peer-to-peer networks. To show the potential of the proposed approach, a cognitive engine for solving topology mismatch problem in unstructured peer-topeer networks will be presented. To evaluate the proposed approach, computer simulations have been conducted using the cognitive engine designed for solving topology mismatch problem and then the results are compared with the results obtained for two existing algorithms called PROP-O and X-BOT for solving topology mismatch problem. It has been shown that the proposed cognitive engine performs better than the existing algorithms with respect to end-to-end delays and delays of mismatched paths.