Research On Dynamic Spectrum Access And Routing Algorithm For Cognitive Radios

In recent years, cognitive radio (CR) has emerged as a new technology in wireless communication. It can sense idle spectrum in communication environment, and enables unlicensed users to have the ability of intelligently identifying and using variable spectrum by reorganizing wireless network systems. So CR can solve the problem of scarcer and scarcer wireless spectrum resource and the fact of underutilization of licensed spectrum in wireless communication. Research results from home and abroad show that CR has high spectrum efficiency and represents a new development trend in the future so that CR has become an important research topic. The dissertation mainly researches dynamic spectrum access and routing algorithm for cognitive radio networks, in depth with the support of the Program from National Science Fundation of China. The work in the dissertation can be summarized as follows:Firstly, a media access control (MAC) technology, with the theory of stopping rule to decide the spectrum sensing problem, is proposed for cognitive radio wireless networks. The secondary (unlicensed) users contend the opportunity to access the spectrum by exchanging control frames. Then the secondary users, which acquire the opportunity successfully, select the optimal number of channels to sense according to the theory of stopping rule, and use idle channels within them. An analytic model is proposed to analyze the optimal number of channels and the throughput of our proposed MAC scheme quantitatively. The simulation results show that our proposed scheme increases the throughput significantly, up to 34.1% at least, campared to the scheme in which a fixed number of channels are sensed.Secondly, a novel MAC scheme, utilizing the optimal idle licensed channel with channel reservation, is proposed for cognitive radio wireless networks. The scheme enables the secondary users to select the optimal idle licensed channel based on spectrum sensing of multiple licensed channels and reserve some channels. When the licensed users want to use channels occupied by secondary users, secondary users must vacant them, and reconstruct the communication links in reservation channels. So the secondary users will not interfere with licensed users, and the communication links affected by licensed users will not be lost. The scheme also solves the problems of hidden terminals and exposed terminals. An analytic model is proposed to analyze the throughput of our proposed MAC scheme under the saturated network case quantitatively. The simulation results show that our proposed MAC scheme increases the throughput up to 129.6%, campared to the scheme without channel reservation.Finally, a novel routing protocol, based on the spectrum-tree, is proposed for cognitive radio wireless networks. The protocol enables the secondary users to select the optimal unused licensed channel after sensing a number of licensed channels, and establish a spectrum-tree in the optimal channel. To reduce the loading of root in a spectrum-tree, some users, which sense the same number of idle channels, are selected as roots. So multiple roots form multiple spectrum-trees which have small nodes, and the loading of each root can be reduced. By using the spectrum-trees and a new route metric about delay, the route, with an efficient route recovery method, can be established for the secondary users. The simulation results show that our proposed routing protocol shortens the average end-to-end delay significantly with increasing control overhead slightly.