认知无线电中序贯检测算法和遗传算法的研究
|
摘要
随着无线通信技术的发展及应用,频谱资源日趋紧缺。解决这一问题的有效途径是采用认知无线电技术对频谱资源实现再利用,以提高频谱效率。本论文主要进行了以下两个方面的研究：频谱检测过程中恶意用户的剔除和参数配置过程中遗传算法的应用。
频谱共享的基础是找到可用频谱,从而频谱检测的准确性是至关重要的。当前的频谱感知技术主要分为辅助频谱感知和独立频谱感知两种,独立频谱感知主要包括发射机信号检测、接收机信号检测、协作检测等检测技术。其中协作检测联合多个本地CR用户共享检测信息,由聚合中心完成空闲频谱的判定,可以获得更准确的检测结果。但是CRN中可能存在恶意用户以及检测准确度较低的用户,恶意用户以一定方式发送错误频谱感知信息,来自于以上两者的错误信息会影响聚合中心对信道占用检测的准确度。本文将恶意用户剔除功能和认知用户权值引入到了序贯检测算法,根据用户检测统计结果进行恶意用户剔除可以排除恶意用户的影响,认知用户权值的引入对用户检测结果进行加权计算可以提高检测准确度。仿真结果表明,提出的改进算法所需的采样点数较少,检测性能有一定提高,并且随着恶意用户数量的增加,检测准确度仍比较稳定。
为了在不同的通信环境以及用户的使用需求下获得更好的性能,无线终端参数配置要在多个参数组合间进行权衡,即需要使用多目标优化算法得到最优参数配置方案。遗传算法模拟生物进化中模式进行搜索,可以实现多目标优化功能。但是遗传算法在进化过程中比较容易陷入局部最优解,为了克服这一缺陷,在遗传算法中引入了适应度函数尺度变换和自适应交叉、变异概率以降低陷入局部最优解的概率并提高进化收敛速度。仿真结果表明,基于改进遗传算法的参数配置方案相较原方案,能够获得性能更好的参数配置组合。虽然其在进化过程中仍然可能会陷入停滞的状态,但经过几代进化后能够跳出这种状态,最终得到满足要求的配置方案。
With the development and application of the communication technology, serious rare resource problem is occurred in spectrum allocation. CR technology could provide some solution for the above problem. This thesis is focused on the following two problems:the removing of malicious terminals and the weight of all CR terminals in SPRT for spectrum detection and the application of genetic algorithm in parameter configuration.
Spectrum sharing technology is based on the detection result of the available spectrum, so the accuracy of the detection is quite important in CR. The cooperative detection which joints the terminals'sensing information at the aggregation center can achieve more accurate results. However, if malicious CR terminals which transmit error sensing information or the CR terminals which cannot sense the spectrum accurately exist in CRN, the detection result in aggregation center may be incorrect. Consequently, a detection algorithm which can reduce the impact is necessary. SPRT is improved by the mechanism of removing the malicious terminals and weighting all the CR terminals. Simulation results show that the improved SPRT can achieve high detection accuracy which needs fewer samples. Meanwhile, it can obtain robust performance with the increase of the malicious users.
The CR terminals maybe adapted to different communication environment and different applications as intelligent terminals, which must own the ability of finding the best combination of different parameters. Genetic algorithm which simulates the phase-out mode in biological evolution can searches the solution space effectively. Whereas general genetic algorithm has inherent defects, it tends to convergent to local optimal solution. Fitness function scale transformation is introduced to overcome this drawback with the adaptive crossover and mutation probability. Thereafter, we can reduce the probability of falling into local optimal solution and improve the speed of evolution convergence. The improved algorithm can achieve better parameter combination and it can jump out this situation after several generations of evolution especially fall into the state of stagnation.
频谱共享的基础是找到可用频谱,从而频谱检测的准确性是至关重要的。当前的频谱感知技术主要分为辅助频谱感知和独立频谱感知两种,独立频谱感知主要包括发射机信号检测、接收机信号检测、协作检测等检测技术。其中协作检测联合多个本地CR用户共享检测信息,由聚合中心完成空闲频谱的判定,可以获得更准确的检测结果。但是CRN中可能存在恶意用户以及检测准确度较低的用户,恶意用户以一定方式发送错误频谱感知信息,来自于以上两者的错误信息会影响聚合中心对信道占用检测的准确度。本文将恶意用户剔除功能和认知用户权值引入到了序贯检测算法,根据用户检测统计结果进行恶意用户剔除可以排除恶意用户的影响,认知用户权值的引入对用户检测结果进行加权计算可以提高检测准确度。仿真结果表明,提出的改进算法所需的采样点数较少,检测性能有一定提高,并且随着恶意用户数量的增加,检测准确度仍比较稳定。
为了在不同的通信环境以及用户的使用需求下获得更好的性能,无线终端参数配置要在多个参数组合间进行权衡,即需要使用多目标优化算法得到最优参数配置方案。遗传算法模拟生物进化中模式进行搜索,可以实现多目标优化功能。但是遗传算法在进化过程中比较容易陷入局部最优解,为了克服这一缺陷,在遗传算法中引入了适应度函数尺度变换和自适应交叉、变异概率以降低陷入局部最优解的概率并提高进化收敛速度。仿真结果表明,基于改进遗传算法的参数配置方案相较原方案,能够获得性能更好的参数配置组合。虽然其在进化过程中仍然可能会陷入停滞的状态,但经过几代进化后能够跳出这种状态,最终得到满足要求的配置方案。
With the development and application of the communication technology, serious rare resource problem is occurred in spectrum allocation. CR technology could provide some solution for the above problem. This thesis is focused on the following two problems:the removing of malicious terminals and the weight of all CR terminals in SPRT for spectrum detection and the application of genetic algorithm in parameter configuration.
Spectrum sharing technology is based on the detection result of the available spectrum, so the accuracy of the detection is quite important in CR. The cooperative detection which joints the terminals'sensing information at the aggregation center can achieve more accurate results. However, if malicious CR terminals which transmit error sensing information or the CR terminals which cannot sense the spectrum accurately exist in CRN, the detection result in aggregation center may be incorrect. Consequently, a detection algorithm which can reduce the impact is necessary. SPRT is improved by the mechanism of removing the malicious terminals and weighting all the CR terminals. Simulation results show that the improved SPRT can achieve high detection accuracy which needs fewer samples. Meanwhile, it can obtain robust performance with the increase of the malicious users.
The CR terminals maybe adapted to different communication environment and different applications as intelligent terminals, which must own the ability of finding the best combination of different parameters. Genetic algorithm which simulates the phase-out mode in biological evolution can searches the solution space effectively. Whereas general genetic algorithm has inherent defects, it tends to convergent to local optimal solution. Fitness function scale transformation is introduced to overcome this drawback with the adaptive crossover and mutation probability. Thereafter, we can reduce the probability of falling into local optimal solution and improve the speed of evolution convergence. The improved algorithm can achieve better parameter combination and it can jump out this situation after several generations of evolution especially fall into the state of stagnation.
引文
[1]Kolodzy P. Spectrum Policy Task Force: Findings And Recommendations[J]. International Symposium On Advanced Radio Technologies (ISART), March2003
[2]Mchenry M. Report On Spectrum Occupancy Measurements. Spectrum Occupancy Measurments, http://www.sharedspectrum.com/?section=nsf_summary
[3]Mitola J et al. Cognitive Radios:Making Software Radios More Personal[C]. IEEE Personal Communications,1999,6(4)
[4]J. Mitola. Cognitive Radio: An Integrated Agent Architecture For Software Defined Radio. Phd Dissertation, Royal Inst. Technol. (KTH), Stockholm, Sweden,2000.
[5]Project Summary. Nets-Prowin: High Performance Cognitive Radio Platform With Integrated Physical And Networklayer Capabilities. WINLAB, Rutgers University
[6]Brodersen R. W., Wolisz A., And Cabric D., et al. CORVUS:A Cognitive Radio Approach For Usage Of Virtual Unlicensed Spectrum. White Paper Berkeley Wireless Research Center. UC Berkeley, Ju1.29,2004
[7]Project Summary. Nets-Prowin: Cognitive Radio For Open Access To Spectrum. WINLAB, Rutgers University.
[8]Weiss T. And Jondral F. Spectrum Pooling: An Innovative Strategy For The Enhancement Of Spectrum Efficiency[C]. IEEE Communications Magazine. Mar,2004,42:8-13.
[9]Mengtuc And Jondral F. Extracting The Channel allocation Information In A Spectrum Pooling System Using A Profiler Delay And Multiply Nonlinearity[C]. Proceedings Of The 2003 IEEE Workshop On Statistical Signal Processing.Sept.28-Oct.1,2003, St. Louis, MO, USA
[10]Cabric G., Demestichas P.And Grandblaise D., et al. Cognitive Radio, Spectrum And Radio Resource Management[C]. Wireless World Research Forum Working Group 6 White Paper, 2004
[11]Digham, F.F., M.S. alouini, And M.K. Simon. On The Energy Detection Of Unknown Signals Over Fading Channels[C]. In Communications,2003. ICC' 03. IEEE International Conference On.2003.
[12]Alayesh, M.A. And N. Ghani. Cognitive Radios Power Control Game Under Rayleigh Slow Fading Channel[C]. In Communications And Information Technology (ICCIT),2011 International Conference On.2011.
[13]Alam, S., et al. Further Results On Area Under The ROC Curve Of Energy Detectors Over Generalized Fading Channels[C]. In Sarnoff Symposium,2011 34th IEEE.2011
[14]Cabric, D., S.M. Mishra, And R.W. Brodersen. Implementation Issues In Spectrum Sensing For Cognitive Radios[C]. In Signals, Systems And Computers,2004. Conference Record Of The Thirty-Eighth Asilomar Conference On.2004.
[15]Cabric, D., A. Tkachenko, And R.W. Brodersen. Spectrum Sensing Measurements Of Pilot, Energy, And Collaborative Detcction[C]. In Military Communications Conference,2006. MILCOM 2006. IEEE.2006.
[16]周贤伟,王建萍,王春江.认知无线电[M].北京：国防工业出版社,2008.1
[17]Ghozzi, M., et al. Cyclostatilonarilty-Based Test For Detection Of Vacant Frequency Bands[C]. Tn Cognitive Radio Oriented Wireless Networks And Communications,2006.1 st International Conference On.2006.
[18]Brown T X. An Analysis of Unlicensed Device Operation in Licensed Broadcast Sercice Bands[C]. In Proc. IEEE DySPAN 2005, November,2005:11-29
[19]Zhao Qing,Sadler B.M. A Surveu Of Dynamic Spectrum Access[J]. Signal Processing Magazine, IEEE.2007,24(3):79-89
[20]王悦,冯春燕,曾志民,郭彩丽.认知无线电网络中基于合作的频谱检测研究[J].电信科学.2008,24(7)：54-57
[21]Cabric D, Mishra S M, Brodcrsen R W. Implementation Issues In Spectrum Sensing For Cognitive Radios. In: Asilomar Conference On Signals, Systems, And Computers, Asilomar, CA, Nov 2004
[22]王小平,曹立明.遗传算法——理论、应用于软件实现[M].西安：西安交通大学出版社,2002.68-86
[23]牛利民.移动计算设备电能消耗估计[D].合肥：中国科学技术大学,2008
[24]Qusay. Cognitive Networks:Towards Self-Aware Networks[M].Wiley,2007:368
[25]Ruiliang, C, P. Jung-Min, And B. Kaigui. Robust Distributed Spectrum Sensing In Cognitive Radio Networks[C]. In INFOCOM 2008.2008:1876-7884.
[26]Cao Lili, Zheng Haitao. Distributed Spectrum allocation Via Local Bargaining[C]. IEEE SECON, Santa Clara, California, USA, Sept 2005
[27]Z. Tian And B. M. Sadler, Weighted Energy Detection Of Ultra Wideband Signals[C]. In IEEE 6th Workshop On Signal Processing Advances In Wireless Communications, Pp.1068 - 1072, June 2005.
[28]徐鑫.认知无线电网络中频谱感知技术研究[D].北京：北京交通大学：2011
[29]J. F. Hauris, Donya He, Geremy Michel, Cahit Ozbay. Cognitive Radio And RF Communications Design Optimization Using Genetic algorithms[J]. Military Communications Conference,2007. MILCOM 2007
[30]周小飞,张宏纲.认知无线电原理及应用[M].北京邮电大学出版社.2007：17-38,99-132
[31]Zhao Zhijin, Peng Zhen, Zheng Shilian, et al. Cognitive Radio Spectrum Allocation Using Evolutionary Algorithms[J], IEEE Transactions On Wireless Communications, 2009.8(9):4421-4425
[32]Rieser CJ. Biologically Inspired Cognitive Radio Engine Model Utilizing Distributed Genetic Radio Engine Model Utilizing Distributed Genetic Algorithm[C]. Secure and Robust Wireless Communication and Networking,2004
[33]D. E. Goldberg. Genetic Algorithms in Search, Optimization, and Machine Lcarning[M]. Addison-Wesley Publishing Co.,1989
[34]郭彩丽,张天魁,曾志民,冯春燕.认知无线电关键技术及应用的研究现状[J].电信科学.2006,(8)：50-55
[35]黄晨,魏胜群,张华伟徐婷.基于遗传算法的快速认知无线电系统[J].现代电子技术,2010,33(7)：10-13
[36]张静,周正,高万鑫.基于二进制量子粒子群算法的认知无线电决策引擎[J].仪器仪表学报,2011,32(2)：451-456.
[37]王国强,李金龙,张敏.多目标遗传算法求解认知无线电性能优化问题[J].计算机工程与应用,2007,43(20)：159-162.
[38]Si Chen, Newman, T.R.,Evans, J.B.,Wyglinski, A.M.. Genetic Algorithm-Based Optimization For Cognitive Radio Networks[J]. Sarnoff Symposium,2010 IEEE.2010,4.1-6
[39]赵知劲,郑仕链,尚俊娜等.基于量子遗传算法的认知无线电决策引擎研究[J].物理学报.2007,56(11)：6760-6766
[40]Zhenyu Zhang, Xiaoyao Xie.Application Research Of Evolution In Cognitive Radio Based On GA[J].In:Proc.IEEE.2008.1575-1579
[41]任子武,伞冶.自适应遗传算法的改进及在系统辨识中应用研究[J].系统仿真学报.2006,18(1)：41-66
[42]Hisao Ishibuchi. A Multi-Objective Genetic Local Search Algorithm And Its Application To Flowshop Scheduling. IEEE transcations on system, man, and cybernetics,1998.28(3):392-403
[43]苏传生.基于智能混合优化算法的不规则件优化排样技术研究[D].青岛：青岛科技大学,2009
[2]Mchenry M. Report On Spectrum Occupancy Measurements. Spectrum Occupancy Measurments, http://www.sharedspectrum.com/?section=nsf_summary
[3]Mitola J et al. Cognitive Radios:Making Software Radios More Personal[C]. IEEE Personal Communications,1999,6(4)
[4]J. Mitola. Cognitive Radio: An Integrated Agent Architecture For Software Defined Radio. Phd Dissertation, Royal Inst. Technol. (KTH), Stockholm, Sweden,2000.
[5]Project Summary. Nets-Prowin: High Performance Cognitive Radio Platform With Integrated Physical And Networklayer Capabilities. WINLAB, Rutgers University
[6]Brodersen R. W., Wolisz A., And Cabric D., et al. CORVUS:A Cognitive Radio Approach For Usage Of Virtual Unlicensed Spectrum. White Paper Berkeley Wireless Research Center. UC Berkeley, Ju1.29,2004
[7]Project Summary. Nets-Prowin: Cognitive Radio For Open Access To Spectrum. WINLAB, Rutgers University.
[8]Weiss T. And Jondral F. Spectrum Pooling: An Innovative Strategy For The Enhancement Of Spectrum Efficiency[C]. IEEE Communications Magazine. Mar,2004,42:8-13.
[9]Mengtuc And Jondral F. Extracting The Channel allocation Information In A Spectrum Pooling System Using A Profiler Delay And Multiply Nonlinearity[C]. Proceedings Of The 2003 IEEE Workshop On Statistical Signal Processing.Sept.28-Oct.1,2003, St. Louis, MO, USA
[10]Cabric G., Demestichas P.And Grandblaise D., et al. Cognitive Radio, Spectrum And Radio Resource Management[C]. Wireless World Research Forum Working Group 6 White Paper, 2004
[11]Digham, F.F., M.S. alouini, And M.K. Simon. On The Energy Detection Of Unknown Signals Over Fading Channels[C]. In Communications,2003. ICC' 03. IEEE International Conference On.2003.
[12]Alayesh, M.A. And N. Ghani. Cognitive Radios Power Control Game Under Rayleigh Slow Fading Channel[C]. In Communications And Information Technology (ICCIT),2011 International Conference On.2011.
[13]Alam, S., et al. Further Results On Area Under The ROC Curve Of Energy Detectors Over Generalized Fading Channels[C]. In Sarnoff Symposium,2011 34th IEEE.2011
[14]Cabric, D., S.M. Mishra, And R.W. Brodersen. Implementation Issues In Spectrum Sensing For Cognitive Radios[C]. In Signals, Systems And Computers,2004. Conference Record Of The Thirty-Eighth Asilomar Conference On.2004.
[15]Cabric, D., A. Tkachenko, And R.W. Brodersen. Spectrum Sensing Measurements Of Pilot, Energy, And Collaborative Detcction[C]. In Military Communications Conference,2006. MILCOM 2006. IEEE.2006.
[16]周贤伟,王建萍,王春江.认知无线电[M].北京：国防工业出版社,2008.1
[17]Ghozzi, M., et al. Cyclostatilonarilty-Based Test For Detection Of Vacant Frequency Bands[C]. Tn Cognitive Radio Oriented Wireless Networks And Communications,2006.1 st International Conference On.2006.
[18]Brown T X. An Analysis of Unlicensed Device Operation in Licensed Broadcast Sercice Bands[C]. In Proc. IEEE DySPAN 2005, November,2005:11-29
[19]Zhao Qing,Sadler B.M. A Surveu Of Dynamic Spectrum Access[J]. Signal Processing Magazine, IEEE.2007,24(3):79-89
[20]王悦,冯春燕,曾志民,郭彩丽.认知无线电网络中基于合作的频谱检测研究[J].电信科学.2008,24(7)：54-57
[21]Cabric D, Mishra S M, Brodcrsen R W. Implementation Issues In Spectrum Sensing For Cognitive Radios. In: Asilomar Conference On Signals, Systems, And Computers, Asilomar, CA, Nov 2004
[22]王小平,曹立明.遗传算法——理论、应用于软件实现[M].西安：西安交通大学出版社,2002.68-86
[23]牛利民.移动计算设备电能消耗估计[D].合肥：中国科学技术大学,2008
[24]Qusay. Cognitive Networks:Towards Self-Aware Networks[M].Wiley,2007:368
[25]Ruiliang, C, P. Jung-Min, And B. Kaigui. Robust Distributed Spectrum Sensing In Cognitive Radio Networks[C]. In INFOCOM 2008.2008:1876-7884.
[26]Cao Lili, Zheng Haitao. Distributed Spectrum allocation Via Local Bargaining[C]. IEEE SECON, Santa Clara, California, USA, Sept 2005
[27]Z. Tian And B. M. Sadler, Weighted Energy Detection Of Ultra Wideband Signals[C]. In IEEE 6th Workshop On Signal Processing Advances In Wireless Communications, Pp.1068 - 1072, June 2005.
[28]徐鑫.认知无线电网络中频谱感知技术研究[D].北京：北京交通大学：2011
[29]J. F. Hauris, Donya He, Geremy Michel, Cahit Ozbay. Cognitive Radio And RF Communications Design Optimization Using Genetic algorithms[J]. Military Communications Conference,2007. MILCOM 2007
[30]周小飞,张宏纲.认知无线电原理及应用[M].北京邮电大学出版社.2007：17-38,99-132
[31]Zhao Zhijin, Peng Zhen, Zheng Shilian, et al. Cognitive Radio Spectrum Allocation Using Evolutionary Algorithms[J], IEEE Transactions On Wireless Communications, 2009.8(9):4421-4425
[32]Rieser CJ. Biologically Inspired Cognitive Radio Engine Model Utilizing Distributed Genetic Radio Engine Model Utilizing Distributed Genetic Algorithm[C]. Secure and Robust Wireless Communication and Networking,2004
[33]D. E. Goldberg. Genetic Algorithms in Search, Optimization, and Machine Lcarning[M]. Addison-Wesley Publishing Co.,1989
[34]郭彩丽,张天魁,曾志民,冯春燕.认知无线电关键技术及应用的研究现状[J].电信科学.2006,(8)：50-55
[35]黄晨,魏胜群,张华伟徐婷.基于遗传算法的快速认知无线电系统[J].现代电子技术,2010,33(7)：10-13
[36]张静,周正,高万鑫.基于二进制量子粒子群算法的认知无线电决策引擎[J].仪器仪表学报,2011,32(2)：451-456.
[37]王国强,李金龙,张敏.多目标遗传算法求解认知无线电性能优化问题[J].计算机工程与应用,2007,43(20)：159-162.
[38]Si Chen, Newman, T.R.,Evans, J.B.,Wyglinski, A.M.. Genetic Algorithm-Based Optimization For Cognitive Radio Networks[J]. Sarnoff Symposium,2010 IEEE.2010,4.1-6
[39]赵知劲,郑仕链,尚俊娜等.基于量子遗传算法的认知无线电决策引擎研究[J].物理学报.2007,56(11)：6760-6766
[40]Zhenyu Zhang, Xiaoyao Xie.Application Research Of Evolution In Cognitive Radio Based On GA[J].In:Proc.IEEE.2008.1575-1579
[41]任子武,伞冶.自适应遗传算法的改进及在系统辨识中应用研究[J].系统仿真学报.2006,18(1)：41-66
[42]Hisao Ishibuchi. A Multi-Objective Genetic Local Search Algorithm And Its Application To Flowshop Scheduling. IEEE transcations on system, man, and cybernetics,1998.28(3):392-403
[43]苏传生.基于智能混合优化算法的不规则件优化排样技术研究[D].青岛：青岛科技大学,2009