基于认知的无线网络自适应通信方法
|
摘要
现代无线通信网络正朝着宽带化、数字化、智能化的方向发展。其中，如何在军事对抗、抢险救灾与应急通信等复杂环境条件下，使通信网络自适应周围动态变化环境，进而对所承载业务提供通信保障是广泛关注的问题，也是本文的研究目标。其中，复杂环境是指在空域、时域及频域范围内，影响系统通信保障能力的环境因素多而杂，且这些因素相互作用影响，难于分析。
认知无线电技术凭借其“通信参数灵活可调整”及“动态频谱伺机接入”等特点，被认为是可以有效实现上述目标的无线通信技术之一，并被众多专用网络或授权网络所接受，成为重要的通信辅助手段。然而，复杂环境下频谱资源的不确定性，使得利用上述认知技术实现网络自适应通信依然面临许多技术问题。为此，本文在分析国内外文献的基础上，从资源的角度与业务传输的角度，研究了资源时变对业务传输的影响，并从QoS保证、低干扰碰撞概率以及低路由中断概率等几方面，具体研究了以下几个问题：
第一，分层分布式网络体系架构及业务传输模型。目前，分层分布式网络架构凭借其分级编配及规模易扩展等特点，被广泛应用于具有层级指挥特点的网络中，如军事战术网、应急指挥网等。因此，本文首先分析了分层分布式网络体系架构，并讨论了该架构下的自适应功能机制。之后，结合上述网络架构及自适应机制，讨论了复杂环境下的信道可用性及链路可用性，以及分层分布式网络架构下的IP分组业务传输特性。
第二，面向聚合业务QoS约束的接纳控制方法。复杂环境下的资源不确定性，将导致业务流在传输及汇聚过程中表现出不同程度的“碎片化”特点，并严重影响业务聚合节点处的排队性能。为此本文建立了汇聚节点处缓存区长度及分配通道速率与丢包率及排队时延间的复杂函数关系。通过建立起来的函数关系，结合多目标优化方法，研究了面向丢包率及排队时延联合约束的接纳控制方法。该方法在保证丢包率的前提下，能够进一步保证排队时延的约束。
第三，基于概率预测的动态信道自适应切换方法。针对复杂环境下，干扰类型的多样化，本文从统计规律上将复杂环境下干扰类型分类为：连续、周期及随机等三种类型，并进一步研究了不同干扰样式下的信道空闲状态持续时长概率区间预测方法。其突出特点是能够给出预测误差的估计值，从而利用区间预测的能力减小了预测误差带来的负面影响。并以概率预测结果为依据，通过设定概率选择门限，指导具体的动态信道分配及自适应切换，这一方面可以使节点提前退出当前信道以避免干扰到来时产生的碰撞，另一方面可以使节点合理选择切换目标信道以降低切换概率。
第四，基于链路可用性的自适应路由选择方法。首先结合网络拓扑模型及信道可用性及链路可用性模型，研究复杂环境下基于链路可用性的路由尺度。并以该路由尺度为判据，通过改进蚁群路由选择算法，研究了复杂环境下的路由选择方法，以降低收发两端所建立路径上的业务阻断概率。其中，重点考虑了信道和链路的时变因素与可靠性因素，及其导致的收发两端所建立路径上的信道切换次数和信道切换率问题，通过选择可靠性较高的单跳路由形成端到端路由，从而避免了不必要的链路重构与路由重构。
Modern wireless communication networks are moving towards broadband, information-digitizing and intelligent. In which, how to make network be adaptive to the dynamically changing environment and provide communications supporting the carried packet data service in military confrontation, disaster reliefemergency and communications, is a wide spread concern.
Cognitive radio technology has the "adjustable and flexible communication parameters " and "opportunistic dynamic spectrum access" features, is considered to be an effective solution to the above problem of wireless communication technology. However, using cognitive radio technology in complex environment to achieve adaptive communication still faces many technical challenges. Therefore, on the basis of the analysis of domestic and foreign literature, the present dissertation research the resource time-varying influence on the traffic flow from the angle of resources and traffic propagation. And specific research the following problem from the aspects of QoS-guarantee, lower collision probability and lower routes disruption probability:
First, hierarchical distributed network architecture and model of service transmission. At present，with the features of clustering and easy expansibility, the hierarchical and distributed network structure is widely used in the networks which has the character of hierarchical command, for example, the tactical networks and the emergency network. Therefore, the present dissertation first analysis the hierarchical distributed network architecture. And then, discussed the adaptive function mechanism with this framework, discussed the channel availability and link availability in this complex environment, as well as IP characteristics of packet service transmission with hierarchical distributed network architecture.
Second, admission control method facing aggregation service QoS constraints. Complex environment will lead to uncertainty spectrum resources, and will lead to the fragments of traffic flow in transmitting and aggregating process, and has important influence on aggregation service queueing performance. Therefore, this dissertation building up the functional relationship between buffer-size service-rate packet loss ratio and queuing time-delay. And by the use of multi-objective optimization method research the admission control method facing the QoS constraint of packet loss rate and queuing delay.
Third, dynamic channel adaptive switching method based on probability prediction. For the diversification of interference type in complex environment, this dissertation from the statistical law for duration type classified the interference as: continuous, periodic and random several kinds, and study a prediction method of channel idle state’s duration probability under a variety of styles interval. Its outstanding feature is the ability to give an estimate of the prediction error, thus take advantage of the ability of interval prediction to reduce the negative impact of the prediction error. And based on the results of probability prediction, which leads specific dynamic channel allocation and adaptive switching by setting the threshold of probability selection. On one hand, this can make node early exit current channel to avoid the collision generating during the arrival of interference. On the other hand, nodes can make a reasonable choice of switching target channel to reduce the switching probability.
Forth, adaptive routing method based on link ability. First, research the link usability under multi-channel combination, and further research routing metric based on link availability under complex environment. And treat this routing metric as criterion, using adaptive routing algorithm with global dynamic optimization study demand routing selection method in complex environments, in order to reduce service blocking probability on the path. Among them, time-varying factors and reliability factors of channel and link are under serious consideration, and the channel switching times and channel switching frequency problem on communication path of both transmitter and receiver which caused by it. In the aspect of routing selection method, selecting the link with high reliability on statistical significance to form route, thereby avoiding unnecessary link reconstruction and route reconstruction.
认知无线电技术凭借其“通信参数灵活可调整”及“动态频谱伺机接入”等特点，被认为是可以有效实现上述目标的无线通信技术之一，并被众多专用网络或授权网络所接受，成为重要的通信辅助手段。然而，复杂环境下频谱资源的不确定性，使得利用上述认知技术实现网络自适应通信依然面临许多技术问题。为此，本文在分析国内外文献的基础上，从资源的角度与业务传输的角度，研究了资源时变对业务传输的影响，并从QoS保证、低干扰碰撞概率以及低路由中断概率等几方面，具体研究了以下几个问题：
第一，分层分布式网络体系架构及业务传输模型。目前，分层分布式网络架构凭借其分级编配及规模易扩展等特点，被广泛应用于具有层级指挥特点的网络中，如军事战术网、应急指挥网等。因此，本文首先分析了分层分布式网络体系架构，并讨论了该架构下的自适应功能机制。之后，结合上述网络架构及自适应机制，讨论了复杂环境下的信道可用性及链路可用性，以及分层分布式网络架构下的IP分组业务传输特性。
第二，面向聚合业务QoS约束的接纳控制方法。复杂环境下的资源不确定性，将导致业务流在传输及汇聚过程中表现出不同程度的“碎片化”特点，并严重影响业务聚合节点处的排队性能。为此本文建立了汇聚节点处缓存区长度及分配通道速率与丢包率及排队时延间的复杂函数关系。通过建立起来的函数关系，结合多目标优化方法，研究了面向丢包率及排队时延联合约束的接纳控制方法。该方法在保证丢包率的前提下，能够进一步保证排队时延的约束。
第三，基于概率预测的动态信道自适应切换方法。针对复杂环境下，干扰类型的多样化，本文从统计规律上将复杂环境下干扰类型分类为：连续、周期及随机等三种类型，并进一步研究了不同干扰样式下的信道空闲状态持续时长概率区间预测方法。其突出特点是能够给出预测误差的估计值，从而利用区间预测的能力减小了预测误差带来的负面影响。并以概率预测结果为依据，通过设定概率选择门限，指导具体的动态信道分配及自适应切换，这一方面可以使节点提前退出当前信道以避免干扰到来时产生的碰撞，另一方面可以使节点合理选择切换目标信道以降低切换概率。
第四，基于链路可用性的自适应路由选择方法。首先结合网络拓扑模型及信道可用性及链路可用性模型，研究复杂环境下基于链路可用性的路由尺度。并以该路由尺度为判据，通过改进蚁群路由选择算法，研究了复杂环境下的路由选择方法，以降低收发两端所建立路径上的业务阻断概率。其中，重点考虑了信道和链路的时变因素与可靠性因素，及其导致的收发两端所建立路径上的信道切换次数和信道切换率问题，通过选择可靠性较高的单跳路由形成端到端路由，从而避免了不必要的链路重构与路由重构。
Modern wireless communication networks are moving towards broadband, information-digitizing and intelligent. In which, how to make network be adaptive to the dynamically changing environment and provide communications supporting the carried packet data service in military confrontation, disaster reliefemergency and communications, is a wide spread concern.
Cognitive radio technology has the "adjustable and flexible communication parameters " and "opportunistic dynamic spectrum access" features, is considered to be an effective solution to the above problem of wireless communication technology. However, using cognitive radio technology in complex environment to achieve adaptive communication still faces many technical challenges. Therefore, on the basis of the analysis of domestic and foreign literature, the present dissertation research the resource time-varying influence on the traffic flow from the angle of resources and traffic propagation. And specific research the following problem from the aspects of QoS-guarantee, lower collision probability and lower routes disruption probability:
First, hierarchical distributed network architecture and model of service transmission. At present，with the features of clustering and easy expansibility, the hierarchical and distributed network structure is widely used in the networks which has the character of hierarchical command, for example, the tactical networks and the emergency network. Therefore, the present dissertation first analysis the hierarchical distributed network architecture. And then, discussed the adaptive function mechanism with this framework, discussed the channel availability and link availability in this complex environment, as well as IP characteristics of packet service transmission with hierarchical distributed network architecture.
Second, admission control method facing aggregation service QoS constraints. Complex environment will lead to uncertainty spectrum resources, and will lead to the fragments of traffic flow in transmitting and aggregating process, and has important influence on aggregation service queueing performance. Therefore, this dissertation building up the functional relationship between buffer-size service-rate packet loss ratio and queuing time-delay. And by the use of multi-objective optimization method research the admission control method facing the QoS constraint of packet loss rate and queuing delay.
Third, dynamic channel adaptive switching method based on probability prediction. For the diversification of interference type in complex environment, this dissertation from the statistical law for duration type classified the interference as: continuous, periodic and random several kinds, and study a prediction method of channel idle state’s duration probability under a variety of styles interval. Its outstanding feature is the ability to give an estimate of the prediction error, thus take advantage of the ability of interval prediction to reduce the negative impact of the prediction error. And based on the results of probability prediction, which leads specific dynamic channel allocation and adaptive switching by setting the threshold of probability selection. On one hand, this can make node early exit current channel to avoid the collision generating during the arrival of interference. On the other hand, nodes can make a reasonable choice of switching target channel to reduce the switching probability.
Forth, adaptive routing method based on link ability. First, research the link usability under multi-channel combination, and further research routing metric based on link availability under complex environment. And treat this routing metric as criterion, using adaptive routing algorithm with global dynamic optimization study demand routing selection method in complex environments, in order to reduce service blocking probability on the path. Among them, time-varying factors and reliability factors of channel and link are under serious consideration, and the channel switching times and channel switching frequency problem on communication path of both transmitter and receiver which caused by it. In the aspect of routing selection method, selecting the link with high reliability on statistical significance to form route, thereby avoiding unnecessary link reconstruction and route reconstruction.
引文
[1] Chiron P, Njedjou E, Seite P, Gosse K. Architectures for IP-basednetwork-assisted mobility management across heterogeneous networks[J].Wireless Communications，15(2)，2008：18-25.
[2] Hall, Mary W, Marina d, Gil Y. Lucas R F. Self-Configuring Applications forHeterogeneous Systems: Program Composition and Optimization UsingCognitive Techniques[J]. Proceedings of the IEEE，96(5)，2008：849-862.
[3] Mitola J, Maguire Q. Cognitive radio: making software radios morepersonal[J]. Personal Communications，6(4)，1999：13-18.
[4] Yongkang L, Xuemin S, Hongwei L. Deploying cognitive cellular networksunder dynamic resource management[J]. Wireless Communications，20(2)，2013：3386-3407.
[5] Cummings M, Heath S. Mode switching and software download for softwaredefined radio: the SDR Forum approach[J]. Communications Magazine，37(8)，1999：104-106.
[6] Kone V, Santa B, Yang L, Yang X, Zhao B. The Effectiveness ofOpportunistic Spectrum Access: A Measurement Study[J]. Networking,IEEE/ACM Transactions，20(6)，2012：2005-2016.
[7] Frank H, Fitzek P, Marcos D. Cognitive wireless networks[M]. Netherlands:Springer Press，2007：657-678.
[8] Fangmin Xu, Luyong Zhang, Zheng Zhou, Yabin Ye. Spectrum awarelocation-based routing in cognitive UWB network[C]. Proceedings of the3rdInternational Conference on Cognitive Radio oriented Wireless Networks andCommunications (CrownCom)，Singapore，IEEE，2008：1-5.
[9] Krishnamurthy S, Thoppian M, Venkatesan S, Prakash R. Control channelbased MAC-layer configuration routing and situation awareness for cognitiveradio networks[A]. Proceedingsof Military Communications Conference(MILCOM)[C]，Atlantic City: IEEE，2005：455-460.
[10] Cheng G, Liu W, Li Y, Cheng W. Joint on-demand routing and spectrumassignment in cognitive radio networks[C]. Proceedings of IEEEInternational Conference on Communications (ICC)，IEEE Press，2007：6499-6503.
[11] Cheng G, Liu W, Li Y, Cheng W. Spectrum aware on-demand routing incognitive radio networks[C]. Proceedings of the2nd IEEE InternationalSymposium on NewFrontiers in Dynamic Spectrum Access Networks(DySPAN)，IEEE Press，2007：571-574.
[12] Huisheng M, Li-li Z, Xiao M, Yongjian L. Spectrum aware routing formulti-hop cognitive radio networks with a single transceiver[C]. Proceedingsof Third International Conference on Cognitive Radio Oriented WirelessNetworks and Communications (CrownCom)，Singapore，IEEE Press，2008：1-6.
[13] Stevenson C, Chouinard G, Zhongding L, Wendong H. IEEE802.22: Thefirst cognitive radio wireless regional area network standard[J].Communications Magazin，47(1)，2009：130-138.
[14] Cicconetti C, Erta A, Lenzini L, Mingozzi E. Performance Evaluation of theIEEE802.16MAC for QoS Support[J]. Mobile Computing, IEEETransactions，6(1)，2007：26-38.
[15] Manoj B S, Rao R R, Zorzi M. CogNet: a cognitive complete knowledgenetwork system[J]. Wireless Communications, IEEE，15(6)，2008：81-88.
[16] Seelig F W. A Description of the August2006XG Demonstrations at FortA.P. Hill[C]. DySPAN2007,2nd IEEE International Symposium，2007：1-12.
[17] Niyato D, Hossain E. Cognitive radio for next-generation wireless networks:an approach to opportunistic channel selection in ieee802.11-based wirelessmesh[J]. Wireless Communications，16(1)，2009.
[18] Mitola J. Cognitive radio: An integrated agent architecture for softwaredefined radio[D]. PhD Thesis，Royal Inst of Technology (KTH)，2000.
[19] Haykin S. Cognitive radio: brain-empowered wireless communications[J].IEEE Sel Areas Commun，23(2)，2005：201-220.
[20] Federal Communications Commission, Spectrum policy task force report[R].http://www.fcc.gov/sptf/files/SEWGFinalReport_1.pdf. Accessed Nov2002.
[21] Facilitating opportunities for flexible, efficient and reliable spectrum useemploying cognitive radio technologies in Federal CommunicationsCommission Spectrum Policy Task Force[R].FCC，Washington, DC，2005：103-108.
[22] Jondral F K. Software-defined radio-basics and evolution to cognitiveradio[J]. EURA SIP J Wirel Commun Netw，2005.
[23] International Telecommunication Union. Definitions of software definedradioand cognitive radio system[R]. Report ITU-R SM, SMSeries, Spectrummanagement，2009.
[24] Jouini W, Ernst D, Moy C, Palicot J. Multi-armed bandit based policies forcognitive radio’s decision making issues[C]. Proceedings of the3rdinternational conference on Signals. Circuits and Systems (SCS)，2009：1-6.
[25] Palicot J. Radio Engineering: From Software Radio to Cognitive Radio[M].Wiley, UK，2011.
[26] Hatfield D, Weiser P. Property rights in spectrum: Taking the next step[C].20051st IEEE International Symposium on New Frontiers in DynamicSpectrum Access Networks，DySPAN，2005：43-55.
[27] Dawei C, Sixing Y. Mining Spectrum Usage Data: a Large-scale SpectrumMeasurement Study[C]. MobiCom’09，Sept.20-25，2009：13-24.
[28] Willkomm D, Machiraju S, Bolot J, Wolisz A. Primary user behavior incellular networks and implications for dynamic spectrum access[J]. IEEECommunications Magazine，47(3)，2009：88-95.
[29] Willkomm D, Machiraju S, Bolot J, Wolisz A. Primary users in cellularnetworks: a large-scale measurement study[C].20083rd IEEE Symposium onNew Frontiers in Dynamic Spectrum AccessNetworks，DySPAN，2008：401-411.
[30] Su H, Zhang X. Cross-layer based opportunistic MAC protocols for QoSprovisionings over cognitive radio wireless networks[J]. IEEE Journal onSelected Area in Communications，26(1)，2008：118-129.
[31] Wellens M, Baynast A, Mahone P. Exploiting historical spectrum occupancyinformation for adaptive spectrum sensing[C]. IEEE WCNC，2008：17-722.
[32] Kaneko S, Nomoto S, Ueda T, Nomura S, Takeuchi K. Predicting radioresource availability in cognitive radio-An experimental examination[C].3rd International Conference on Cognitive Radio Oriented Wireless Networksand Communications，CrownCom2008，Singapore，2008.
[33] Takeuchi K, Kaneko S, Nomoto S. Radio environment prediction forcognitive radio[C].3rd International Conference on Cognitive Radio OrientedWireless Networks and Communications，CrownCom2008，Singapore，2008.
[34] Kaneko S, Nomoto S, Ueda T, Nomura S, Takeuchi K, Sugiyama K.Experimental verification on the prediction of the trend in radio resourceavailability in cognitive radio[C].2007IEEE66th Vehicular TechnologyConference，VTC2007-Fall，2007：1568-1572.
[35] Tevfik Yucek. Huseyin Arslan. A Survey of Spectrum Sensing Algorithmsfor Cognitive Radio Applications[J]. Communications Survey&Tutorials，IEEE Volume11, Issue1, First Quarter2009. pp:116-130.
[36] Geirhofer S, Lang T, Sadler B M. A measurement-based model for dynamicspectrum access in WLAN channels[C]. Military CommunicationsConference2006，2006：1-7.
[37] Zhao Q, Tong L. An anthram Swami. Decentralized cognitive mac fordynamic spectrum access[C]. DySPAN2005，2005：224-232.
[38] Stefan G, Tong L, Brian M. Cognitive Radios For Dynamic Spectrum Accessin the Time Domain: Modeling and Exploiting White Space[J]. IEEECommunications Magazine，45(5)，2007：66-72.
[39] Pattanayak S, Nandi R. Identification of spectrum holes using ANN model forcognitive radio applications[C]. EUROCON,2013IEEE，2013：133-137.
[40] Jinghua G, Hong J, Yi L, Xi L. A novel spectrum handoff managementscheme based on SVM in cognitive radio networks[C]. Communications andNetworking in China (CHINACOM),20116th International ICSTConference，2011：645-649.
[41] Jones S, Merheb N, Wang I. An experiment for sensing based opportunisticspectrum access in CSMA/CA networks[C]. DySPAN.05，IEEECommunications Society，2005：593-596.
[42] Akbar I A, Tranter W H. Dynamic spectrum allocation in cognitive radiousing hidden markov models: Poisson distributed case[C].2007IEEESoutheast Con，2007：196-201.
[43] Yarkan S, Arslan H. Binary time series approach to spectrum prediction forcognitive radio[C]. IEEE66th Vehicular Technology Conference, VTC2007-Fall，2007：1563-1567.
[44] Wen Z, Luo T, Xiang W, Majhi S, Ma Y. Autoregressive spectrum holeprediction model for cognitive radio systems[C]. ICC2008-2008IEEEInternational Conference on Communications Workshops，2008：154-157.
[45] Xiaoshuang X, Tao J, Wei C, Yan H. Spectrum prediction in cognitive radionetworks[J]. Wireless Communications，20(2)，2013：90-96.
[46] Luo T, Jiang T, Xiang W, Chen H-H. A subcarriers allocation scheme forcognitive radio systems based on multi-carrier modulation[J]. IEEETransactions on Wireless Communications，7(9)，2008：3335-3340.
[47] Chen Z, Robert C. Predietion of channel state for cognitive radio usinghigher-order hidden markov model[C]. IEEE Southeast Con2010proceedings，2010：276-282.
[48] Zhao Q, Tong L, Yunxia Chen. Decentralized cognitive MAC foropportunistic spectrum access in Ad-Hoc networks: A POMDP Framework[J].IEEE Journal on Communications，25(3)，2007：589-600.
[49] Zhigang W, Tao L, etal. Auto regressive spectrum hole prediction model forcognitive radio systems[C]. ICC workshop，08Proceedings，2008：154-157.
[50] Lei G, Shoufeng D, Wanbin T, Shaoqian L. Anti-intermittence source routingprotocol in distributed cognitive radio network[C].2008InternationalConference on Wireless Communications Networking and Mobile Computing,WiCOM2008, Dalian, China,2008.
[51]苏曦，沈树群，冯志永，陈星.认知无线电系统中的自适应动态频谱分配方法[J].电子与信息学报，31(12)，2009：2801-2806.
[52] Jo O, Cho D-H. Efficient spectrum matching based on spectrumcharacteristics in cognitive radio systems[C].7th Annual WirelessTelecommunications Symposium, WTS2008，2008：230-235.
[53] Jo O, Park J H, Cho D-H. QoS provisioning spectrum management based onintelligent matching and reservation for cognitive radio system[C].2ndInternational Conference on Cognitive Radio Oriented Wireless Networksand Communications, CrownCom，2007：530-534.
[54] Liang Y, Zeng Y, Hoang A. Sensing-throughput trade off for cognitive radionetworks[J]. Wireless Communications, IEEE Transactions，7(4)，2008：1326-1337.
[55] Yulong Z, Yudong Y, Baoyu Z. Outage probability analysis of cognitivetransmissions: Impact of spectrum sensing overhead[J]. WirelessCommunications, IEEE Transactions，9(8)，2010：2676–2688.
[56] Cormio C, Chowdhury KR. A survey on MAC protocols for cognitive radionetworks[J]. Ad Hoc Netw，7(7)，2009：1315-1329.
[57] Akyildiz IF, Lee W, Chowdhury K. CRAHNs: cognitive radio ad hocnetworks[J]. Ad Hoc Netw，7(5)，2009：810-836.
[58] Wu S, Lin C, Tseng Y, Sheu J. A new multi-channel MAC protocol withon-demand channel assignment for multi-hop mobile ad hoc networks[C].IEEE DySPAN，2005：203-213.
[59] Ma L, Han X, Shen C-C. Dynamic open spectrum sharing for wireless ad hocnetworks[C]. IEEE DySPAN，2005：203-213.
[60] Ma L, Shen C-C, Ryu B. Signal-radio adaptive channel algorithm forspectrum agile wireless ad hoc networks[C]. IEEE DySPAN，2007：547-558.
[61] Yoo S, Nan H, Hyon T. DCR-MAC: distributed cognitive radio MACprotocol for wireless ad hoc networks[J]. Wirel Commun Mobile Comput，9(5)，2009：631-653.
[62] Timmers M, Pollin S, Dejonghe A. A distributed multi-channel MACprotocol for multi-hop cognitive radio networks[J]. IEEE Trans VehTechnol，59(1)，2010：446-459.
[63] Yin S, Chen D, Zhang Q, Li S. Prediction-based throughput optimization fordynamic spectrum access[J]. IEEE Trans Veh Technol，60(3)，2011：1284-1289.
[64] Cordeiro C, Challapali K. C-MAC: a cognitive MAC protocol formulti-channel wireless networks[C]. IEEE DySPAN，2007：147-157.
[65] Wang Y, Ren P, Wu G. A throughput-aimed MAC protocol with QoSprovision for cognitive ad hoc networks[J]. IEICE Trans Commun，E93-B(6)，2010：1426-1429.
[66] Su H, Zhang X. Cross-layer based opportunistic MAC protocols for QoSprovisionings over cognitive radio wireless networks[J]. IEEE J Sel AreasCommun，26(1)，2008：118-129.
[67] Iyer G, Lim Y. Efficient multi-channel MAC protocol and channel allocationschemes for TDMA based cognitive radio networks[C]. ICCSP，2011：394-398.
[68] Benedetto M, Nardis L. Cognitive routing in UWB networks[C]. Proceedingsof IEEE International Conference on Ultra-Wideband. IEEE Press，2006：381-386.
[69] Thomas R, DaSilva L, Marathe M, Wood K. Critical design decisions forcognitive networks[C]. Proceedings of IEEE International Conference onCommunications (ICC)，IEEE Press，2007：3993-3998.
[70] Gabriella M, Benedetto Di, Nardis L. Cognitive routing models in UWBnetworks[A]. Proceedings of the3rd International Conference on CognitiveRadio Oriented Wireless Networks and Communications (CrownCom)[C]，IEEE Press，2008：1-6.
[71] Miao M, Tsang D. Joint spectrum sharing and fair routing in cognitive radionetworks[A]. Proceedings of Consumer Communications and NetworkingConference (CCNC)[C]，IEEE Press，2008：978-982.
[72] Liu Y, Grace D. Improving capacity for wireless Adhoc communicationsusing cognitive routing[A]. Proceedings of the3rd International Conferenceon Cognitive Radio Oriented Wireless Networks and Communications(CrownCom)[C]，Singapore，IEEE，2008：1-6.
[73] Chang W, Mikio P. Minimum weight routingbased on a common link controlradio for cognitive wireless Ad-hoc networks[A]. Proceedings ofInternational Conferenceon Wireless Communications and MobileComputing[C]，ACM Press，2007：399-404.
[74] Tao C, Honggang Z, Maggio G M, Chlamtac I. CogMesh: a cluster basedcognitive radio network[C]. Proceedings of the2nd IEEE InternationalSymposium on New Frontiers in Dynamic Spectrum Access Networks(DySPAN)，2007：168-178.
[75] Tao C, Honggang Z, Maggio G, Chlamtac I. Topology management inCogMesh: a cluster-based cognitive radiomesh network[A]. Proceedings ofIEEE International Conference on Communications (ICC)[C]，IEEE，2007：6516-6521.
[76] Baldo N, Zorzi M. Fuzzy logic for cross-layer optimization in cognitive radionetworks[J]. IEEE Communications Magazine，46(4)，2008：64-71.
[77] Hang S, Xi Z. Cross-layer based opportunistic MAC protocols for QoSprovisionings over cognitive radio wireless networks[J]. IEEE Journal onSelected Areas in Communications，26(1)，2008：118-129.
[78] Saeed R, Khatun S, Ali B, Abdullah K. Ultra-wide band interferencemitigation using cross-layer cognitive radio[A]. Proceedings of IFIPInternational Conference on Wireless and Optical CommunicationsNetworks[C]，Bangalore，IEEEPress，2006：1-5.
[79] Shakkottai S, Rappaport S, Karlsson P. Cross-layer design for wirelessnetworks[J]. IEEE Communications Magazine，41(10)，2003：74-80.
[80] Srivastava V, Motani M. Cross-layer design: a survey and the road ahead[J].IEEE Communications Magazine，43(12)，2005：112-119.
[81] Foukalas F, Gazis V, Alonistioti N. Cross-layer design proposals for wirelessmobile networks: a survey and taxonomy[J]. IEEE Communications Surveys&Tutorials，10(1)，2008：70-85.
[82] Wang Q, Zheng H. Route and spectrum selection in dynamic spectrumnetworks[C]. Proceedings of the3rd IEEE Consumer Communications andNetworking Conference (CNCC)，IEEE Press，2006：625-629.
[83] Xin C. A novel layered graph model for topology formation and routing indynamic spectrum access networks[A]. Proceedings of IEEE InternationalDynamic Spectrum Access Networks (DySPAN)[C]，IEEE Press，2005：308-317.
[84] Vali D, Paskalis S, Merakos L, Kaloxylos A. A survey of internet QoSsignaling[J]. Communications Surveys&Tutorials，6(4)，2004，32-43.
[85] Xiaofei C, Nagaraj S V. Entropy-based spectrum sensing in cognitiveradio[C]. Wireless Telecommunications Symposium,2008. WTS2008，2008：57-61.
[86] Fehske A, Gaeddert J, Reed J H. A new approach to signal classificationusing spectralcorrelation and neural networks[C].20051st IEEE InternationalSymposium on New Frontiers in DynamicSpectrum Access Networks,DySPAN，2005：144-150.
[87] Zhi Q, Shuguang C, Sayed A. Optimal Linear Cooperation for SpectrumSensing in Cognitive Radio Networks[J]. Selected Topics in SignalProcessing，2(1)，2008：28-40.
[88] Sahai A, Hoven N, Tandra R. Some fundamental limits on cognitive radio[C].Allerton Conference on Communication Control and Computing2004，2004：1-7.
[89] Urkowitz H. Energy detection of unknown deterministic signals[J].Proceedings of IEEE，55(4)，1967：523-531.
[90] Wellens M, Wu J, Mallonen P. Evaluation of spectrum occupancy in indoorand outdoor scenario in the context of cognitive radio[C]. IEEE Crown Com2007，2007：420-427.
[91] Khatkhate A, Ray A, Keller E, Gupta S. Symbolic time-series analysis foranomaly detection in mechanical systems[J]. IEEE/ASME Transactions onMechatronics，11(4)，2006：439-447.
[92] Tsybakov B, Georganas N D. Self-Similar Processes in CommunicationsNetworks[J]. IEEE Transactions on Information Theory，44(5)，1998：1713-1725.
[93] Taqqu M S, Willinger W, Robert S. Proof of a Fundamental Result inSelf-similar Traffic Modeling[J]. ACM Computer Communication Review，(27)2，1997：5-23.
[94] Willinger W, Taqqu M S, Sherman R, Wilson D V. Self-similarity throughhigh-variability: statistical analysis of Ethernet LAN traffic at the sourcelevel[J]. IEEE/ACM Trans，Networking，5，1997：71-86.
[95] Perrin E,Harba R, Berzin-Joseph C, Iribarren I. nth-order fractional Brownianmotion and fractional Gaussian noises[J]. Transactions on Signal Processing，49(5)，2001：1049-1059.
[96] Leland W, Taqqu M, Willinger W, Wilson D. On the self-similar nature ofEthernet traffic(extended version)[J]. IEEE/ACM Trans. Networking，2(1)，1994：1-15.
[97] Willinger W, Paxson V, Taqqu M S. Self-similarity and heavytails: structurelmodeling of network traffic[C]. Staticstical Techniques and Applications, Eds.Cambridge, MA: Birkh user，1998：27-53.
[98] Pacheco J, Roman D. A Tool for Long-Range Dependent Analysis via the R/SStatistic. CIC '06.15th International Conference on Computing，2006：361-366.
[99] Shi J, Zhu H. Merging and Splitting Self-Similar Traffic[C]. FifthAsia-Pacific Conference on Communications，1999：110-114.
[100] Enachescu M, Ganjali Y, Goel A, McKeown N, Roughgarden T. Routers withvery small buffers[C]. IEEE INFOCOM2006，2006：1-11.
[101] Aggarwal A, Savage S, Anderson T. Understanding the performance of TCPpacing[C]. IEEE INFOCOM，2000：1157-1165.
[102] Huang C, et al. Fast simulation of self-similar traffic in ATM networks[C].ICC’95，1995：438-444.
[103] Dhamdhere A, Dovrolis C. Open issues in router buffer sizing[J]. ACMSIGCOMM Computer Communications Review，36(1)，2006：87-92.
[104] Neidhardt A, Wang J L. The concept of relevant time scales and itsapplication to queuing analysis of self-similar traffic[C]. ACM SIGMETRICS1998，1998：222-232.
[105] Likhanov N, Tsybakov B, Georganas ND. Analysis of an ATM buffer withself-similar (―Fractal‖) input traffic[C]. IEEE INFOCOM’95，Washington:IEEE Computer Society，1995：985-992.
[106] Grossglauser M, Bolot J. On the relevance of long-range dependence innetwork traffic[J]. IEEE/ACM Transactions on Networking，7(5)，1999：629-640.
[107] Stathis C, Maglaris B. Modeling the self-similar behavior of networktraffic[J]. Computer Network，34(1)，2000：37-47.
[108] Erramilli A, Roughan M. Self-similar traffic and network dynamics[J].Proceedings of the IEEE，90(5)，2002：800-819.
[109]李景聪，李正斌.自相似业务量下网络的排队性能分析[J].北京大学学报，38(5)，2002(9).
[110] Duffield N G, O'Connell N. Large deviations and overflow probabilities forthe general single-server queue with application[J]. Math Proc CambridgePhilos Soc，9，1995：363-375.
[111] Rockafellar R T. Convex Analysis[M]. Princeton University Press，1970：251-258.
[112] A H柯尔莫戈洛夫.函数论与泛函分析初步[M].高等教育出版社，2006.
[113]李庆扬等.现代数值分析[M].高等教育出版社，1995.
[114]郑金华.多目标进化算法及其应用[M]．北京：科学出版社，2007.
[115] Goldberg D E. A comparative analysis of selection schemes used in geneticalgorithms[C]. Foundations of Genetic Algorithms, SanMateo, CA: MorganKaufmann，1991：69-93.
[116] Kalyanmoy Deb, Amrit Pratap, Sameer Agarwal, et al. A fast and elitistmulti-objective genetic algorithm: NSGA-II[J]. IEEE Transactions onEvolutionary Computation，6(2)，2002：182-197.
[117] Tipping M E. Sparse Bayesian Learning and the Relevance VectorMachine[J]. Machine learning Research，2001，6(1)：211-244.
[118] Camps V G, Martinez R M. Nonlinear System Identification With CompositeRelevance Vector Machines[J]. IEEE Signal Processing Letters，2007，14(4)：279-282.
[119] Ranjan P．Efficient Routing Algorithms for Multi-Channel DynamicSpectrum Access Networks[C]. New Frontiers in Dynamic Spectrum AccessNetworks，2007．
[120] Cheng G, Liu W. Li Y, Cheng W. Spectrum Aware On-Demand Routing inCognitive Radio Networks[C]. New Frontiers in Dynamic Spectrum AccessNetworks，2007．
[121] Wang X, Wong A, Ho PH. Stochastic Medium Access for Cognitive RadioAd-Hoc Networks[J]. IEEE Journal on Selected Areas in Communications，29(4)，2011：770-783.
[2] Hall, Mary W, Marina d, Gil Y. Lucas R F. Self-Configuring Applications forHeterogeneous Systems: Program Composition and Optimization UsingCognitive Techniques[J]. Proceedings of the IEEE，96(5)，2008：849-862.
[3] Mitola J, Maguire Q. Cognitive radio: making software radios morepersonal[J]. Personal Communications，6(4)，1999：13-18.
[4] Yongkang L, Xuemin S, Hongwei L. Deploying cognitive cellular networksunder dynamic resource management[J]. Wireless Communications，20(2)，2013：3386-3407.
[5] Cummings M, Heath S. Mode switching and software download for softwaredefined radio: the SDR Forum approach[J]. Communications Magazine，37(8)，1999：104-106.
[6] Kone V, Santa B, Yang L, Yang X, Zhao B. The Effectiveness ofOpportunistic Spectrum Access: A Measurement Study[J]. Networking,IEEE/ACM Transactions，20(6)，2012：2005-2016.
[7] Frank H, Fitzek P, Marcos D. Cognitive wireless networks[M]. Netherlands:Springer Press，2007：657-678.
[8] Fangmin Xu, Luyong Zhang, Zheng Zhou, Yabin Ye. Spectrum awarelocation-based routing in cognitive UWB network[C]. Proceedings of the3rdInternational Conference on Cognitive Radio oriented Wireless Networks andCommunications (CrownCom)，Singapore，IEEE，2008：1-5.
[9] Krishnamurthy S, Thoppian M, Venkatesan S, Prakash R. Control channelbased MAC-layer configuration routing and situation awareness for cognitiveradio networks[A]. Proceedingsof Military Communications Conference(MILCOM)[C]，Atlantic City: IEEE，2005：455-460.
[10] Cheng G, Liu W, Li Y, Cheng W. Joint on-demand routing and spectrumassignment in cognitive radio networks[C]. Proceedings of IEEEInternational Conference on Communications (ICC)，IEEE Press，2007：6499-6503.
[11] Cheng G, Liu W, Li Y, Cheng W. Spectrum aware on-demand routing incognitive radio networks[C]. Proceedings of the2nd IEEE InternationalSymposium on NewFrontiers in Dynamic Spectrum Access Networks(DySPAN)，IEEE Press，2007：571-574.
[12] Huisheng M, Li-li Z, Xiao M, Yongjian L. Spectrum aware routing formulti-hop cognitive radio networks with a single transceiver[C]. Proceedingsof Third International Conference on Cognitive Radio Oriented WirelessNetworks and Communications (CrownCom)，Singapore，IEEE Press，2008：1-6.
[13] Stevenson C, Chouinard G, Zhongding L, Wendong H. IEEE802.22: Thefirst cognitive radio wireless regional area network standard[J].Communications Magazin，47(1)，2009：130-138.
[14] Cicconetti C, Erta A, Lenzini L, Mingozzi E. Performance Evaluation of theIEEE802.16MAC for QoS Support[J]. Mobile Computing, IEEETransactions，6(1)，2007：26-38.
[15] Manoj B S, Rao R R, Zorzi M. CogNet: a cognitive complete knowledgenetwork system[J]. Wireless Communications, IEEE，15(6)，2008：81-88.
[16] Seelig F W. A Description of the August2006XG Demonstrations at FortA.P. Hill[C]. DySPAN2007,2nd IEEE International Symposium，2007：1-12.
[17] Niyato D, Hossain E. Cognitive radio for next-generation wireless networks:an approach to opportunistic channel selection in ieee802.11-based wirelessmesh[J]. Wireless Communications，16(1)，2009.
[18] Mitola J. Cognitive radio: An integrated agent architecture for softwaredefined radio[D]. PhD Thesis，Royal Inst of Technology (KTH)，2000.
[19] Haykin S. Cognitive radio: brain-empowered wireless communications[J].IEEE Sel Areas Commun，23(2)，2005：201-220.
[20] Federal Communications Commission, Spectrum policy task force report[R].http://www.fcc.gov/sptf/files/SEWGFinalReport_1.pdf. Accessed Nov2002.
[21] Facilitating opportunities for flexible, efficient and reliable spectrum useemploying cognitive radio technologies in Federal CommunicationsCommission Spectrum Policy Task Force[R].FCC，Washington, DC，2005：103-108.
[22] Jondral F K. Software-defined radio-basics and evolution to cognitiveradio[J]. EURA SIP J Wirel Commun Netw，2005.
[23] International Telecommunication Union. Definitions of software definedradioand cognitive radio system[R]. Report ITU-R SM, SMSeries, Spectrummanagement，2009.
[24] Jouini W, Ernst D, Moy C, Palicot J. Multi-armed bandit based policies forcognitive radio’s decision making issues[C]. Proceedings of the3rdinternational conference on Signals. Circuits and Systems (SCS)，2009：1-6.
[25] Palicot J. Radio Engineering: From Software Radio to Cognitive Radio[M].Wiley, UK，2011.
[26] Hatfield D, Weiser P. Property rights in spectrum: Taking the next step[C].20051st IEEE International Symposium on New Frontiers in DynamicSpectrum Access Networks，DySPAN，2005：43-55.
[27] Dawei C, Sixing Y. Mining Spectrum Usage Data: a Large-scale SpectrumMeasurement Study[C]. MobiCom’09，Sept.20-25，2009：13-24.
[28] Willkomm D, Machiraju S, Bolot J, Wolisz A. Primary user behavior incellular networks and implications for dynamic spectrum access[J]. IEEECommunications Magazine，47(3)，2009：88-95.
[29] Willkomm D, Machiraju S, Bolot J, Wolisz A. Primary users in cellularnetworks: a large-scale measurement study[C].20083rd IEEE Symposium onNew Frontiers in Dynamic Spectrum AccessNetworks，DySPAN，2008：401-411.
[30] Su H, Zhang X. Cross-layer based opportunistic MAC protocols for QoSprovisionings over cognitive radio wireless networks[J]. IEEE Journal onSelected Area in Communications，26(1)，2008：118-129.
[31] Wellens M, Baynast A, Mahone P. Exploiting historical spectrum occupancyinformation for adaptive spectrum sensing[C]. IEEE WCNC，2008：17-722.
[32] Kaneko S, Nomoto S, Ueda T, Nomura S, Takeuchi K. Predicting radioresource availability in cognitive radio-An experimental examination[C].3rd International Conference on Cognitive Radio Oriented Wireless Networksand Communications，CrownCom2008，Singapore，2008.
[33] Takeuchi K, Kaneko S, Nomoto S. Radio environment prediction forcognitive radio[C].3rd International Conference on Cognitive Radio OrientedWireless Networks and Communications，CrownCom2008，Singapore，2008.
[34] Kaneko S, Nomoto S, Ueda T, Nomura S, Takeuchi K, Sugiyama K.Experimental verification on the prediction of the trend in radio resourceavailability in cognitive radio[C].2007IEEE66th Vehicular TechnologyConference，VTC2007-Fall，2007：1568-1572.
[35] Tevfik Yucek. Huseyin Arslan. A Survey of Spectrum Sensing Algorithmsfor Cognitive Radio Applications[J]. Communications Survey&Tutorials，IEEE Volume11, Issue1, First Quarter2009. pp:116-130.
[36] Geirhofer S, Lang T, Sadler B M. A measurement-based model for dynamicspectrum access in WLAN channels[C]. Military CommunicationsConference2006，2006：1-7.
[37] Zhao Q, Tong L. An anthram Swami. Decentralized cognitive mac fordynamic spectrum access[C]. DySPAN2005，2005：224-232.
[38] Stefan G, Tong L, Brian M. Cognitive Radios For Dynamic Spectrum Accessin the Time Domain: Modeling and Exploiting White Space[J]. IEEECommunications Magazine，45(5)，2007：66-72.
[39] Pattanayak S, Nandi R. Identification of spectrum holes using ANN model forcognitive radio applications[C]. EUROCON,2013IEEE，2013：133-137.
[40] Jinghua G, Hong J, Yi L, Xi L. A novel spectrum handoff managementscheme based on SVM in cognitive radio networks[C]. Communications andNetworking in China (CHINACOM),20116th International ICSTConference，2011：645-649.
[41] Jones S, Merheb N, Wang I. An experiment for sensing based opportunisticspectrum access in CSMA/CA networks[C]. DySPAN.05，IEEECommunications Society，2005：593-596.
[42] Akbar I A, Tranter W H. Dynamic spectrum allocation in cognitive radiousing hidden markov models: Poisson distributed case[C].2007IEEESoutheast Con，2007：196-201.
[43] Yarkan S, Arslan H. Binary time series approach to spectrum prediction forcognitive radio[C]. IEEE66th Vehicular Technology Conference, VTC2007-Fall，2007：1563-1567.
[44] Wen Z, Luo T, Xiang W, Majhi S, Ma Y. Autoregressive spectrum holeprediction model for cognitive radio systems[C]. ICC2008-2008IEEEInternational Conference on Communications Workshops，2008：154-157.
[45] Xiaoshuang X, Tao J, Wei C, Yan H. Spectrum prediction in cognitive radionetworks[J]. Wireless Communications，20(2)，2013：90-96.
[46] Luo T, Jiang T, Xiang W, Chen H-H. A subcarriers allocation scheme forcognitive radio systems based on multi-carrier modulation[J]. IEEETransactions on Wireless Communications，7(9)，2008：3335-3340.
[47] Chen Z, Robert C. Predietion of channel state for cognitive radio usinghigher-order hidden markov model[C]. IEEE Southeast Con2010proceedings，2010：276-282.
[48] Zhao Q, Tong L, Yunxia Chen. Decentralized cognitive MAC foropportunistic spectrum access in Ad-Hoc networks: A POMDP Framework[J].IEEE Journal on Communications，25(3)，2007：589-600.
[49] Zhigang W, Tao L, etal. Auto regressive spectrum hole prediction model forcognitive radio systems[C]. ICC workshop，08Proceedings，2008：154-157.
[50] Lei G, Shoufeng D, Wanbin T, Shaoqian L. Anti-intermittence source routingprotocol in distributed cognitive radio network[C].2008InternationalConference on Wireless Communications Networking and Mobile Computing,WiCOM2008, Dalian, China,2008.
[51]苏曦，沈树群，冯志永，陈星.认知无线电系统中的自适应动态频谱分配方法[J].电子与信息学报，31(12)，2009：2801-2806.
[52] Jo O, Cho D-H. Efficient spectrum matching based on spectrumcharacteristics in cognitive radio systems[C].7th Annual WirelessTelecommunications Symposium, WTS2008，2008：230-235.
[53] Jo O, Park J H, Cho D-H. QoS provisioning spectrum management based onintelligent matching and reservation for cognitive radio system[C].2ndInternational Conference on Cognitive Radio Oriented Wireless Networksand Communications, CrownCom，2007：530-534.
[54] Liang Y, Zeng Y, Hoang A. Sensing-throughput trade off for cognitive radionetworks[J]. Wireless Communications, IEEE Transactions，7(4)，2008：1326-1337.
[55] Yulong Z, Yudong Y, Baoyu Z. Outage probability analysis of cognitivetransmissions: Impact of spectrum sensing overhead[J]. WirelessCommunications, IEEE Transactions，9(8)，2010：2676–2688.
[56] Cormio C, Chowdhury KR. A survey on MAC protocols for cognitive radionetworks[J]. Ad Hoc Netw，7(7)，2009：1315-1329.
[57] Akyildiz IF, Lee W, Chowdhury K. CRAHNs: cognitive radio ad hocnetworks[J]. Ad Hoc Netw，7(5)，2009：810-836.
[58] Wu S, Lin C, Tseng Y, Sheu J. A new multi-channel MAC protocol withon-demand channel assignment for multi-hop mobile ad hoc networks[C].IEEE DySPAN，2005：203-213.
[59] Ma L, Han X, Shen C-C. Dynamic open spectrum sharing for wireless ad hocnetworks[C]. IEEE DySPAN，2005：203-213.
[60] Ma L, Shen C-C, Ryu B. Signal-radio adaptive channel algorithm forspectrum agile wireless ad hoc networks[C]. IEEE DySPAN，2007：547-558.
[61] Yoo S, Nan H, Hyon T. DCR-MAC: distributed cognitive radio MACprotocol for wireless ad hoc networks[J]. Wirel Commun Mobile Comput，9(5)，2009：631-653.
[62] Timmers M, Pollin S, Dejonghe A. A distributed multi-channel MACprotocol for multi-hop cognitive radio networks[J]. IEEE Trans VehTechnol，59(1)，2010：446-459.
[63] Yin S, Chen D, Zhang Q, Li S. Prediction-based throughput optimization fordynamic spectrum access[J]. IEEE Trans Veh Technol，60(3)，2011：1284-1289.
[64] Cordeiro C, Challapali K. C-MAC: a cognitive MAC protocol formulti-channel wireless networks[C]. IEEE DySPAN，2007：147-157.
[65] Wang Y, Ren P, Wu G. A throughput-aimed MAC protocol with QoSprovision for cognitive ad hoc networks[J]. IEICE Trans Commun，E93-B(6)，2010：1426-1429.
[66] Su H, Zhang X. Cross-layer based opportunistic MAC protocols for QoSprovisionings over cognitive radio wireless networks[J]. IEEE J Sel AreasCommun，26(1)，2008：118-129.
[67] Iyer G, Lim Y. Efficient multi-channel MAC protocol and channel allocationschemes for TDMA based cognitive radio networks[C]. ICCSP，2011：394-398.
[68] Benedetto M, Nardis L. Cognitive routing in UWB networks[C]. Proceedingsof IEEE International Conference on Ultra-Wideband. IEEE Press，2006：381-386.
[69] Thomas R, DaSilva L, Marathe M, Wood K. Critical design decisions forcognitive networks[C]. Proceedings of IEEE International Conference onCommunications (ICC)，IEEE Press，2007：3993-3998.
[70] Gabriella M, Benedetto Di, Nardis L. Cognitive routing models in UWBnetworks[A]. Proceedings of the3rd International Conference on CognitiveRadio Oriented Wireless Networks and Communications (CrownCom)[C]，IEEE Press，2008：1-6.
[71] Miao M, Tsang D. Joint spectrum sharing and fair routing in cognitive radionetworks[A]. Proceedings of Consumer Communications and NetworkingConference (CCNC)[C]，IEEE Press，2008：978-982.
[72] Liu Y, Grace D. Improving capacity for wireless Adhoc communicationsusing cognitive routing[A]. Proceedings of the3rd International Conferenceon Cognitive Radio Oriented Wireless Networks and Communications(CrownCom)[C]，Singapore，IEEE，2008：1-6.
[73] Chang W, Mikio P. Minimum weight routingbased on a common link controlradio for cognitive wireless Ad-hoc networks[A]. Proceedings ofInternational Conferenceon Wireless Communications and MobileComputing[C]，ACM Press，2007：399-404.
[74] Tao C, Honggang Z, Maggio G M, Chlamtac I. CogMesh: a cluster basedcognitive radio network[C]. Proceedings of the2nd IEEE InternationalSymposium on New Frontiers in Dynamic Spectrum Access Networks(DySPAN)，2007：168-178.
[75] Tao C, Honggang Z, Maggio G, Chlamtac I. Topology management inCogMesh: a cluster-based cognitive radiomesh network[A]. Proceedings ofIEEE International Conference on Communications (ICC)[C]，IEEE，2007：6516-6521.
[76] Baldo N, Zorzi M. Fuzzy logic for cross-layer optimization in cognitive radionetworks[J]. IEEE Communications Magazine，46(4)，2008：64-71.
[77] Hang S, Xi Z. Cross-layer based opportunistic MAC protocols for QoSprovisionings over cognitive radio wireless networks[J]. IEEE Journal onSelected Areas in Communications，26(1)，2008：118-129.
[78] Saeed R, Khatun S, Ali B, Abdullah K. Ultra-wide band interferencemitigation using cross-layer cognitive radio[A]. Proceedings of IFIPInternational Conference on Wireless and Optical CommunicationsNetworks[C]，Bangalore，IEEEPress，2006：1-5.
[79] Shakkottai S, Rappaport S, Karlsson P. Cross-layer design for wirelessnetworks[J]. IEEE Communications Magazine，41(10)，2003：74-80.
[80] Srivastava V, Motani M. Cross-layer design: a survey and the road ahead[J].IEEE Communications Magazine，43(12)，2005：112-119.
[81] Foukalas F, Gazis V, Alonistioti N. Cross-layer design proposals for wirelessmobile networks: a survey and taxonomy[J]. IEEE Communications Surveys&Tutorials，10(1)，2008：70-85.
[82] Wang Q, Zheng H. Route and spectrum selection in dynamic spectrumnetworks[C]. Proceedings of the3rd IEEE Consumer Communications andNetworking Conference (CNCC)，IEEE Press，2006：625-629.
[83] Xin C. A novel layered graph model for topology formation and routing indynamic spectrum access networks[A]. Proceedings of IEEE InternationalDynamic Spectrum Access Networks (DySPAN)[C]，IEEE Press，2005：308-317.
[84] Vali D, Paskalis S, Merakos L, Kaloxylos A. A survey of internet QoSsignaling[J]. Communications Surveys&Tutorials，6(4)，2004，32-43.
[85] Xiaofei C, Nagaraj S V. Entropy-based spectrum sensing in cognitiveradio[C]. Wireless Telecommunications Symposium,2008. WTS2008，2008：57-61.
[86] Fehske A, Gaeddert J, Reed J H. A new approach to signal classificationusing spectralcorrelation and neural networks[C].20051st IEEE InternationalSymposium on New Frontiers in DynamicSpectrum Access Networks,DySPAN，2005：144-150.
[87] Zhi Q, Shuguang C, Sayed A. Optimal Linear Cooperation for SpectrumSensing in Cognitive Radio Networks[J]. Selected Topics in SignalProcessing，2(1)，2008：28-40.
[88] Sahai A, Hoven N, Tandra R. Some fundamental limits on cognitive radio[C].Allerton Conference on Communication Control and Computing2004，2004：1-7.
[89] Urkowitz H. Energy detection of unknown deterministic signals[J].Proceedings of IEEE，55(4)，1967：523-531.
[90] Wellens M, Wu J, Mallonen P. Evaluation of spectrum occupancy in indoorand outdoor scenario in the context of cognitive radio[C]. IEEE Crown Com2007，2007：420-427.
[91] Khatkhate A, Ray A, Keller E, Gupta S. Symbolic time-series analysis foranomaly detection in mechanical systems[J]. IEEE/ASME Transactions onMechatronics，11(4)，2006：439-447.
[92] Tsybakov B, Georganas N D. Self-Similar Processes in CommunicationsNetworks[J]. IEEE Transactions on Information Theory，44(5)，1998：1713-1725.
[93] Taqqu M S, Willinger W, Robert S. Proof of a Fundamental Result inSelf-similar Traffic Modeling[J]. ACM Computer Communication Review，(27)2，1997：5-23.
[94] Willinger W, Taqqu M S, Sherman R, Wilson D V. Self-similarity throughhigh-variability: statistical analysis of Ethernet LAN traffic at the sourcelevel[J]. IEEE/ACM Trans，Networking，5，1997：71-86.
[95] Perrin E,Harba R, Berzin-Joseph C, Iribarren I. nth-order fractional Brownianmotion and fractional Gaussian noises[J]. Transactions on Signal Processing，49(5)，2001：1049-1059.
[96] Leland W, Taqqu M, Willinger W, Wilson D. On the self-similar nature ofEthernet traffic(extended version)[J]. IEEE/ACM Trans. Networking，2(1)，1994：1-15.
[97] Willinger W, Paxson V, Taqqu M S. Self-similarity and heavytails: structurelmodeling of network traffic[C]. Staticstical Techniques and Applications, Eds.Cambridge, MA: Birkh user，1998：27-53.
[98] Pacheco J, Roman D. A Tool for Long-Range Dependent Analysis via the R/SStatistic. CIC '06.15th International Conference on Computing，2006：361-366.
[99] Shi J, Zhu H. Merging and Splitting Self-Similar Traffic[C]. FifthAsia-Pacific Conference on Communications，1999：110-114.
[100] Enachescu M, Ganjali Y, Goel A, McKeown N, Roughgarden T. Routers withvery small buffers[C]. IEEE INFOCOM2006，2006：1-11.
[101] Aggarwal A, Savage S, Anderson T. Understanding the performance of TCPpacing[C]. IEEE INFOCOM，2000：1157-1165.
[102] Huang C, et al. Fast simulation of self-similar traffic in ATM networks[C].ICC’95，1995：438-444.
[103] Dhamdhere A, Dovrolis C. Open issues in router buffer sizing[J]. ACMSIGCOMM Computer Communications Review，36(1)，2006：87-92.
[104] Neidhardt A, Wang J L. The concept of relevant time scales and itsapplication to queuing analysis of self-similar traffic[C]. ACM SIGMETRICS1998，1998：222-232.
[105] Likhanov N, Tsybakov B, Georganas ND. Analysis of an ATM buffer withself-similar (―Fractal‖) input traffic[C]. IEEE INFOCOM’95，Washington:IEEE Computer Society，1995：985-992.
[106] Grossglauser M, Bolot J. On the relevance of long-range dependence innetwork traffic[J]. IEEE/ACM Transactions on Networking，7(5)，1999：629-640.
[107] Stathis C, Maglaris B. Modeling the self-similar behavior of networktraffic[J]. Computer Network，34(1)，2000：37-47.
[108] Erramilli A, Roughan M. Self-similar traffic and network dynamics[J].Proceedings of the IEEE，90(5)，2002：800-819.
[109]李景聪，李正斌.自相似业务量下网络的排队性能分析[J].北京大学学报，38(5)，2002(9).
[110] Duffield N G, O'Connell N. Large deviations and overflow probabilities forthe general single-server queue with application[J]. Math Proc CambridgePhilos Soc，9，1995：363-375.
[111] Rockafellar R T. Convex Analysis[M]. Princeton University Press，1970：251-258.
[112] A H柯尔莫戈洛夫.函数论与泛函分析初步[M].高等教育出版社，2006.
[113]李庆扬等.现代数值分析[M].高等教育出版社，1995.
[114]郑金华.多目标进化算法及其应用[M]．北京：科学出版社，2007.
[115] Goldberg D E. A comparative analysis of selection schemes used in geneticalgorithms[C]. Foundations of Genetic Algorithms, SanMateo, CA: MorganKaufmann，1991：69-93.
[116] Kalyanmoy Deb, Amrit Pratap, Sameer Agarwal, et al. A fast and elitistmulti-objective genetic algorithm: NSGA-II[J]. IEEE Transactions onEvolutionary Computation，6(2)，2002：182-197.
[117] Tipping M E. Sparse Bayesian Learning and the Relevance VectorMachine[J]. Machine learning Research，2001，6(1)：211-244.
[118] Camps V G, Martinez R M. Nonlinear System Identification With CompositeRelevance Vector Machines[J]. IEEE Signal Processing Letters，2007，14(4)：279-282.
[119] Ranjan P．Efficient Routing Algorithms for Multi-Channel DynamicSpectrum Access Networks[C]. New Frontiers in Dynamic Spectrum AccessNetworks，2007．
[120] Cheng G, Liu W. Li Y, Cheng W. Spectrum Aware On-Demand Routing inCognitive Radio Networks[C]. New Frontiers in Dynamic Spectrum AccessNetworks，2007．
[121] Wang X, Wong A, Ho PH. Stochastic Medium Access for Cognitive RadioAd-Hoc Networks[J]. IEEE Journal on Selected Areas in Communications，29(4)，2011：770-783.