A Minimum Spanning Tree Based Routing Protocol For Multi-hop And Multi-channel Cognitive Radio

Fixed channel allocation technique, used in traditional wireless network implies to assign planned set of frequency channels to each node of the network. Alternately, Cognitive Radio Network (CRN) permits dynamic spectrum access in multiple channels. Hence, CRN is a lot more efficient in bandwidth utilization than the traditional network. CRN is characterized by the selection of channels for every secondary user from available set of channels at each point in the route. A route is feasible only when there is at least one common channel between each pair of nodes in the route. Theoretical graph methods are utilized in traditional multi-hop networks. However, it fails to model multi-hop Cognitive Radio Networks efficiently and to capture required information for optimal routing. For this reasons, a multi-edge planer graph has been designed where conventional routing protocol like AODV, DSDV can be implemented. In multi-edge graph representation, along with the increased number of nodes and edges in the network, more bandwidth is required to update and maintain the routing table. Hence, the overhead for updating and maintaining these tables will increase and degrade the performance of the network. In this paper, a Minimum Spanning Tree (MST) based routing protocol is employed on the planer graph model which is useful to eliminate redundant links and to prevent possible network loops. In this case, the performance of the network will not be affected with increasing number of nodes. This MST based protocol is validated through simulations. The simulation result depicts the improved percentage of successful routes for single radio and multi radio CRN. It also reduces time complexity than earlier conventional routing protocols.