Cognitive Radio Networks Routing Strategy Research Based On Genetic Algorithm

As the rapid progress of the wireless communication, the spectrum resourceshortage has become more critical. Cognitive radio as an effective measure to solve thisshortage has attracted the researchers’ attention. Cognitive radio reaches themultidimensional multiplexing of the radio spectrum. In CR system cognitive user isallowed to access licensed spectrum if it causes no interference to primary user. So thecognitive radio technology can improve the utilization rate of spectrum resources from anew perspective. Network layer routing technology of CR must be fit for characteristicsof CRN and achieve demand of business. In this paper CRN routing strategy based ongenetic algorithm has been researched，then novel routing metrics and routing protocolwhich are fit for CRN are proposed.First, the background and development of CR are introduced. The new feature ofcognitive radio Ad Hoc networks, including spectrum dynamic, spectrum difference andtopology inconstancy, are given and analyzed, these new problems must be solved byCR routing algorithm.In chapter3of this paper, take full account of the impact of channel capacity andchannel stabilization on routing performance. Then according to spectrum utilizationmechanism of CRN and the ON-OFF model of spectrum, a routing metric basedcapacity meeting rate and channel stabilization is proposed which is used for evaluatingpath from the point of view of CRN.In order to meet the demand of QoS, in chapter4and5, routing strategy based ongenetic algorithm is discussed. With this algorithm some path with better performancewould be discovered from the finite routing information. And based this algorithm, thecontent of message and routing discovery mechanism of on-demand routing protocolDSR are improved. The steps of algorithm and protocol are described. Simulation showthat this protocol not only work correctly in CRN，but also provide GA with moreinformation of topology, so as to discover some better performance path.At last, the routing algorithm proposed in this paper is simulated using theextended NS2tool, the results show that the proposed algorithm has good throughputand packet delivery ratio performances, but the routing overhead is slightly larger.