The Cognitive Radio Network Routing Algorithm Based On Pcnn Research

With the development of radio communications, the growing demand for spectrum resources tensions the frequency resources, has been increasingly prominent. Cognitive Radio Networks has been widespread concerned because it can improve the spectrum utilization. In cognitive radio networks, cognitive users can use spectrums "by chance" through detecting the use of authorized spectrums, in order to solve the problem of low utilization of authorized spectrum. However, the use of authorized spectrum for cognitive users depends primarily on the activities of the primary users, when a cognitive user is using an authorized spectrum band, if a primary user appears, the cognitive user must immediately switch to the other spectrum band, which can not cause any interference to the primary user. In cognitive radio networks, cognitive user’s available spectrum resources are dynamic changed in time, space and the spectrum band, making the network topology and available band for cognitive node has highly dynamic characteristics. This is the biggest difference between cognitive radio networks and traditional wireless networks. The traditional distance metrics wireless routing algorithm is no longer applicable to cognitive radio networks, which requires new routing metrics and algorithms for dynamic changes in the characteristics of the network spectrum resource to meet the requirements of cognitive radio networks.The paper includes the following sections:In part one, we put forward the shortest delay routing selection metrics,according to the cognitive node available band dynamic change characteristics, including the transmission delay and channel switching delay. The simulation results show that compared with traditional routing distance metric, the shortest delay routing selection metric can effectively shorten the delay in the network.In part two, Pulse Coupled Neural Network (PCNN), and its linear improved and simplified model MPCNN is introduced and brought in the solution of cognitive radio network routing, which is called MPSPT. The simulation results show that when the network size is larger, cognitive radio network routing algorithm based on MPCNN has the advantage of low time consumption, comparing to the traditional Dijkstra algorithm on solving shortest path tree.In part three, we analyze the shortest path tree when the network topology changes, and put forward an routing algorithm for dynamic cognitive radio networks based on MPCNN, called DMPSPT. Simulation results show that when the network topology changes, the DMPSPT algorithm than static MPCNN algorithm converges faster, with lower time consumption.