Research Of The Spectrum Sharing Problem In Cognitive Radio Network

At present, broadband of the network, business diversification and mode of theexisting fixed spectrum management lead to the fact that the wireless network is facinggreat challenges, the shortage of the wireless network resources. The dynamicheterogeneous network environment, made the contradiction between the business needsand scarce resources increases, restricting the deployment and operation of current andfuture wireless network. Through intelligent study, reasoning and decision of the wirelessnetwork environment, cognitive radio network achieved the effective sharing andoptimized using of the spectrum resources. It has become the fundamental method ofimproving the utilization of spectrum resources. In dynamic and heterogeneousenvironment, cognitive network not only provides new model of wireless spectrummanagement, but also renders superior service to the users.With the background, this paper considers the problem of spectrum sharing incognitive radio network. Some of earlier research models about spectrum sharing, mostlyconsidered the problem in spectrum sensing, spectrum access and spectrum decision.There were few researches about spectrum mobility. And in the models of spectrummobility, although the literature [1] and [2] established models to study the performance ofspectrum mobility, there was only one pair of cognitive radio users in the network. Theliterature [3] considered multiple-pair of users in the network, but the networkenvironment with licensed channel and unlicensed channel was not concerned. The paper[4] and [5] considered this network environment, but they used proactive approach, whichignored the collision between cognitive radio users and primary users.This paper takes the problem of spectrum mobility as the key point of the research,and the main work of this paper as follows: Firstly, the paper proposes a Markov modelimproved from the literature [3], which is based on time slot. And the model is to describethe performance of spectrum handoff in the network environment with licensed channeland unlicensed channel. Secondly, the paper considers two approaches to access thechannel. The one is cognitive radio user accesses the licensed channel and unlicensedchannel randomly, the other is the user accesses the unlicensed channel firstly, and whenthere is no idle unlicensed channel in the network, it will access the licensed channel. Andthis paper analyzes the performance of the two methods respectively. Thirdly, under thecondition that cognitive radio user is in the licensed channel, the model uses reactiveapproach. When the primary user accesses the channel, the cognitive radio user can notsense the collision immediately. There is time delay, and the paper studied the influence onthe performance of the delay. Lastly, the paper considers the influence of the competition between multiple-pair of users. The paper simulated this model by the parameter such asthe throughput rate, the spectrum handoff rate, and the average spectrum handoff delay tocompare the model in the literature [3]. The simulation result shows that, the performanceof the approaches with unlicensed channel is much better than the one without unlicensedchannel. And among the two approaches with unlicensed channel, the performance of themethod that cognitive radio user accesses the unlicensed channel is better. At last, thepaper makes a summary and analyses the shortcoming of this paper, and then points outthe further research work.