Efficient Traffic Aware Multipath Routing Algorithm in Cognitive Networks

Abstract

Cognitive networks embody a sense of dynamic responsiveness as actions are typically taken in response to changing circumstances and changing resource availability, which use prior and current knowledge gained from the network to take actions with respect to the end-to-end goals of the whole network. According to the cognitive network framework, a multi-path routing algorithm based on traffic prediction model, Efficient Traffic Aware Multi-path Routing (ETAMR) is proposed in cognitive networks. Traffic prediction routing scheme has been investigated with ATPRA [1] that is proposed in previous works. ETAMR considers traffic distribution and traffic load to build a multi-path routing, depending on the prediction model-MMSE to construct the prediction matrix and select the primary route with the shortest delay and lowest traffic load, meanwhile according to the real time traffic load it dynamically triggers the backup paths to avoid congestion and balance the traffic load of the network. Further more, ETAMR is able to adaptively build a multi-path routing scheme of the lowest aggregated traffic load by learning and reasoning scheme. Comparing with current routing algorithms, ETAMR has good performances at load balancing and lower transmission delay, which is validated by the simulation.