Research On Spectrum Access And Spectrum Handoff In Cognitive Radio Networks

With the rapid development of wireless communication technology and the demand of wireless data services,the radio spectrum resources become increasingly scarce for the growing requirement of the data transmission rate.On the other hand,the radio spectrum resources which have been allocated by the traditional fixed spectrum management rules vary degrees of inactivity in time and space.Cognitive radio technology,which allows secondary users(SUs)to occupy the idle licensed spectrum dynamically,is an effective and practical technology to solve the scarcity of spectrum resources and improve the spectrum utilization.In cognitive radio networks(CRNs),SUs sense spectrum bands firstly to detect the idle spectrum bands,and then access the idle ones which are satisfied the requirement of the secondary quality of service(QoS).To guarantee the QoS of primary users(PUs),SUs need to perform spectrum handoff timely when PUs come back.Both spectrum access and spectrum handoff include spectrum sensing process,spectrum analysis and decision process.It is clearly reasonable that the key issues for spectrum access and handoff are how to find the idle spectrum quickly and effectively,how to avoid harmful interference to PUs,and how to get the best performance.Therefore,designing effective spectrum access and handoff strategies is a key research content in CRNs.In this dissertation,we study and investigate the spectrum access and handoff technology in CRNs.The contents of this work are listed as follows:Firstly,the performance analysis models for spectrum access and handoff are studied.We use Markov chain to model the activities of PUs and SUs in a CRN,and evaluate the system performance of different spectrum access and handoff strategies.The continuous time Markov chain(CTMC)models for the immediate spectrum handoff strategy,the spectrum handoff strategy with buffering and the spectrum handoff strategy with channel reservation are analyzed in two cases of perfect spectrum sensing and imperfect spectrum sensing.To verify the feasibility and accuracy of the theoretical CTMC models,discrete event system simulation is established.Based on the CTMC models,the secondary blocking probability,forced termination probability,throughput and the spectrum utilization are analyzed,and the collision probability between PUs and SUs in the imperfect sensing is also analyzed.Simulation results demonstrate the influence of the imperfect sensing to the performance of PUs,and the effect of different spectrum handoff strategies to the performance of SUs.Secondly,the spectrum access and handoff strategy based on cooperative sensing is analyzed.For the hard fusion rule of cooperative sensing,an optimal k solution for the k-out-of-n fusion rule is proposed by jointly optimizing the secondary detection threshold and the value of k to minimize the total cooperative sensing error probability,considering both the consistent and inconsistent received primary signal to noise ratio(SNR)of each SU in a secondary user group(SUG).In addition,the cost of spectrum access and handoff strategy based on cooperative sensing is analyzed,while an optimization problem is also proposed for the cost problem.Thirdly,the spectrum access and handoff strategy based on channel sensing sequence(CSS)is discussed.We focus on the design and optimization of the CSS considering the narrowband sensing based CRNs.First of all,we design the recommended channel sensing sequence(RCSS)for the real time service,aiming to minimize the secondary handoff delay by jointly optimizing the sensing sequence and the number of sensing channels subject to the sensing accuracy and secondary link maintenance constraints.The numerical results disclose that the proposed RCSS scheme can achieve better performance of secondary handoff delay than others when the received SNR of the primary signal on different channels is non-identical.Then,we study the design and optimization scheme of the CSS with adaptive modulation in the Rayleigh fading channels,and propose the maximum energy-efficiency CSS,the maximum throughput CSS and the minimum energy consumption CSS.The CSS design problems are solved by the dynamic programming approach.The numerical results verify that the maximum energy-efficiency CSS can make the tradeoff between the secondary achievable throughput and the secondary overheads effectively.In addition,we propose a hybrid spectrum access strategy to improve the secondary performance,which combines spectrum overlay and spectrum underlay.The hybrid strategy means that the SU accesses by overlay strategy if it detects an idle channel according to the CSS,otherwise it accesses by underlay strategy.Finally,the spectrum access and handoff scheme based on energy harvesting is investigated.The energy harvesting technology is proposed to solve the energy consumption issue in CRNs.Considering the energy half-duplex and the energy causality constraints in energy harvesting CRNs(EHCRNs),we put forward an optimized saving-sensing-transmitting(SST)protocol by exploiting the residual energy of the SU's battery efficiently.We define the normal,sensing and sleeping modes according to the value of the SU's save-ratio.In the normal mode,we maximize the secondary energy utilization by jointly optimizing the save-ratio and transmission power subject to the energy causality and minimum secondary throughput constraints.This problem is solved by first determining the best save-ratio in closed-form and then optimizing the power allocation using Dinkelbach method.The sensing mode is optimized using a similar method.In addition,the implementation algorithm of the optimized SST protocol is also given.The numerical results validate that the proposed scheme can achieve better performance.