Routing Design In Cognitive Wireless Mesh Network Using Layered Graph And Ant Colony Algorithm

Routing design in cognitive wireless mesh network is a hot spot in researches. This project is from NSFC and Important National Science& Technology Specific Projects, and has both theoretical and practical values.Cognitive network uses spectrum in a dynamic way, traditional layered graph needs to get all spectrum usage information of the whole network to form the graph. To eliminate this restriction, we put two kinds of route tables in each node. Via these tables, which correspondents to vertical edges and horizontal edges in layered graph respectively, each node can build a local view of the whole graph, and use this local view to guide routing and interface assignment. As all the local views equal a complete layered graph, we make a distributed layered graph. Then we introduced ant colony algorithm into this distributed layered graph model, using ant colony to update the edges and their values in the graph.In this paper, we transplanted two ant colony algorithms into cognitive wireless mesh network, and analyzed their performance. Then we combined the advantages of both algorithms with further improvement to build a new ant colony algorithm. Simulations demonstrated that the distributed layered graph and ant colony algorithm can adapt well into the dynamic spectrum access network.