无线电信号监测若干关键问题研究与实现
摘要
随着现代通信技术的日新月异,对于无线电管理部门的技术要求也相应地与时俱进。鉴于这种情况,国家无线电监测中心力图对现行无线电监测方法和算法提出具有建设性,切实可行的解决方案,以便可以处理现阶段亟待解决的问题。现今,对于空间无线电信号的监测有两个技术性难题,一是对于信号调制方式自动、高效以及准确的识别;二是对于信号发射源被动、便携以及确切的定位。这两个问题归纳到信号处理的领域,可归类为信号调制方式自动识别和信源被动定位问题。无线电监管涉及到国家的军事和民事安全,而这两个问题又是无线电监管中的技术难题,其重要性不言而喻。
针对调制方式识别领域,其主要方法分为基于极大似然法和基于特征量法两种。研究对象一般为有限几种空时可能混合使用的调制方式信号,并且可以较好的加以区分。但是,现行算法中,缺乏可对多种调制方式信号进行有效识别的研究;并且没有很好的方法用于权衡算法性能以及计算量之间的矛盾;针对被动定位领域,按定位参数分,主要方法为基于时间、基于方向以及基于接收强度三种;按是否发射检测电波分,主要分为主动定位和被动定位两种。现有算法缺少对空间信源非协同被动定位的研究,另外现有算法对定位仪器的性能要求较高,且缺少便携性。
本文广泛调研,深入研究两个问题现行主要研究手段和算法的基础上,立足实际应用,针对现行算法的不足,提出了优化的方案:
首先,采用基于高阶累积量和人工神经网络进行23种调制方式识别,提取信号高阶累积量为特征参数,辅以其他时频特征作为特征参数,并引用径向基函数人工神经网络作为分类器。提出权衡系数,用以权衡算法识别性能与计算量;针对实际采集信号,进行下采样后识别,在保障识别性能的前提下,显著缩小计算量。
其次,采用基于时延估计的被动定位,引入群复相关方法,利用复相关以及后统计的方法,提高整体精度。并致力开发DSP平台下的实用系统,从而在保证定位精度的前提下,降低了对设备的要求,增加系统便携性。
两个算法均已具备最终系统。调制方式自动高效识别系统,对于23种基本调制方式的仿真信号准确识别率达到97%以上,对于10种实际采集到的调制方式识别的正确率超过90%;基于时延估计的被动定位系统,实际环境下,对于带宽大于100KHz信号的定位误差小于30m。通过仿真以及实际环境下的测试结果,验证了两个优化方案的可行性。
With the advances of modern communication technology, the technical requirements for radio management departments rise at the same time. Given this situation, the Radio Management of China is attempting to propose constructive and practical solutions based on the existing radio frequency signals monitoring methods and algorithms. So that they could handle the problems occurred. At present time, there are two technical problems for the signals monitoring of radio frequency from the open space. One is the accurate, automatic and efficient signal modulation classification and the other one is passive, portable and exact locating of the signal sources. Radio management is relative with the national military and civilian security. These two technical issues are of great significant, which could be classified into signal Automatic Modulation Classification (AMC) and passive locating of signals.
For the signals modulation classification issue, there are two major methods including the Maximum Likelihood method and Characteristic based method. The research targets are often limited to several space-time mixed modulation signals which could be classified well. However, the existing algorithms are short of effective classification performance for signals with multi-modulation modes. Moreover, there is no good method to balance the confliction between algorithm performance and computational amount. The passive location methods are classified into three types. According to location parameters, they include time-based, direction-based and receptive-intensity-based methods. There are positive and passive locating methods if classified based on whether to emit detection waves. These methods neglect the study on space signal source non-collaborative passive locating and they have higher requirements on locating equipment and less portability.
This thesis propose improved solutions to the above two issues based on in-depth study of existing methods and algorithms. Our methods have more feasibility and are easier to be implemented for the practical applications.
Firstly, High Order Cumulants (HOC) and Artificial Neural Network (ANN) are used to classify 23 modulation modes. Our methods extract HOC as signal characteristics combined with other time-frequency features and cite Radial Base Function (RBF) ANN as a classifier. In addition, a tradeoff ratio is proposed to measure algorithm classification performance and computational amount. Real signals are classified after down-sampling. The computational amount is greatly reduced on the premise of ensuring classification accuracy performance.
Secondly, we apply time-difference-of-arrival (TDOA)-based estimates to locate the signal sources passively. This method use complex-correlation statistics to improve the system precision. We are attempting to develop practical system on the DSP platform. The equipment requirements are much decreased with increased portability on the premise of ensuring locating performance.
These two algorithms both implemented in the real system. The automatic modulation classification system can classify simulated signals of 23 modulation modes with classify rate of 97%. For real signals with 10 modulation modes, the classify rate is above 90%. The locating precision of passive locating system can reach to 30m for the signals whose bandwidth is larger than 100KHz in real environments. The feasibility of the two proposed algorithms has been testified through simulation and real tests.
针对调制方式识别领域,其主要方法分为基于极大似然法和基于特征量法两种。研究对象一般为有限几种空时可能混合使用的调制方式信号,并且可以较好的加以区分。但是,现行算法中,缺乏可对多种调制方式信号进行有效识别的研究;并且没有很好的方法用于权衡算法性能以及计算量之间的矛盾;针对被动定位领域,按定位参数分,主要方法为基于时间、基于方向以及基于接收强度三种;按是否发射检测电波分,主要分为主动定位和被动定位两种。现有算法缺少对空间信源非协同被动定位的研究,另外现有算法对定位仪器的性能要求较高,且缺少便携性。
本文广泛调研,深入研究两个问题现行主要研究手段和算法的基础上,立足实际应用,针对现行算法的不足,提出了优化的方案:
首先,采用基于高阶累积量和人工神经网络进行23种调制方式识别,提取信号高阶累积量为特征参数,辅以其他时频特征作为特征参数,并引用径向基函数人工神经网络作为分类器。提出权衡系数,用以权衡算法识别性能与计算量;针对实际采集信号,进行下采样后识别,在保障识别性能的前提下,显著缩小计算量。
其次,采用基于时延估计的被动定位,引入群复相关方法,利用复相关以及后统计的方法,提高整体精度。并致力开发DSP平台下的实用系统,从而在保证定位精度的前提下,降低了对设备的要求,增加系统便携性。
两个算法均已具备最终系统。调制方式自动高效识别系统,对于23种基本调制方式的仿真信号准确识别率达到97%以上,对于10种实际采集到的调制方式识别的正确率超过90%;基于时延估计的被动定位系统,实际环境下,对于带宽大于100KHz信号的定位误差小于30m。通过仿真以及实际环境下的测试结果,验证了两个优化方案的可行性。
With the advances of modern communication technology, the technical requirements for radio management departments rise at the same time. Given this situation, the Radio Management of China is attempting to propose constructive and practical solutions based on the existing radio frequency signals monitoring methods and algorithms. So that they could handle the problems occurred. At present time, there are two technical problems for the signals monitoring of radio frequency from the open space. One is the accurate, automatic and efficient signal modulation classification and the other one is passive, portable and exact locating of the signal sources. Radio management is relative with the national military and civilian security. These two technical issues are of great significant, which could be classified into signal Automatic Modulation Classification (AMC) and passive locating of signals.
For the signals modulation classification issue, there are two major methods including the Maximum Likelihood method and Characteristic based method. The research targets are often limited to several space-time mixed modulation signals which could be classified well. However, the existing algorithms are short of effective classification performance for signals with multi-modulation modes. Moreover, there is no good method to balance the confliction between algorithm performance and computational amount. The passive location methods are classified into three types. According to location parameters, they include time-based, direction-based and receptive-intensity-based methods. There are positive and passive locating methods if classified based on whether to emit detection waves. These methods neglect the study on space signal source non-collaborative passive locating and they have higher requirements on locating equipment and less portability.
This thesis propose improved solutions to the above two issues based on in-depth study of existing methods and algorithms. Our methods have more feasibility and are easier to be implemented for the practical applications.
Firstly, High Order Cumulants (HOC) and Artificial Neural Network (ANN) are used to classify 23 modulation modes. Our methods extract HOC as signal characteristics combined with other time-frequency features and cite Radial Base Function (RBF) ANN as a classifier. In addition, a tradeoff ratio is proposed to measure algorithm classification performance and computational amount. Real signals are classified after down-sampling. The computational amount is greatly reduced on the premise of ensuring classification accuracy performance.
Secondly, we apply time-difference-of-arrival (TDOA)-based estimates to locate the signal sources passively. This method use complex-correlation statistics to improve the system precision. We are attempting to develop practical system on the DSP platform. The equipment requirements are much decreased with increased portability on the premise of ensuring locating performance.
These two algorithms both implemented in the real system. The automatic modulation classification system can classify simulated signals of 23 modulation modes with classify rate of 97%. For real signals with 10 modulation modes, the classify rate is above 90%. The locating precision of passive locating system can reach to 30m for the signals whose bandwidth is larger than 100KHz in real environments. The feasibility of the two proposed algorithms has been testified through simulation and real tests.
引文
[1] K. Abed-Meraim, J. F. Cardoso, A. Y. Gorokhov. On subspace methods for blind identification of single-input multiple-output FIR systems, IEEE Trans. Signal Processing. Jan 1997, Vol. 45:1-14.
[2] O. Besson, M. Ghogho, and A. Swami. Parameter estimation for random amplitude chirp signals. IEEE Trans. Signal Processing. Dec. 1999. Vol. 47:3208-3219.
[3] M. Ghogho, A. Swami, and B. Garel. Performance analysis of cyclic statistics for the estimation of harmonics in multiplicative and additive noise. IEEE Trans. Signal Processing. Dec. 1999. Vol.47:3235-3249.
[4] Y. V. Zakharov, V. M. Baronkin, and T. C. Tozer. DFT-based frequency estimators with narrow acquisition range. Proc. Inst. Elect. Eng. Commun. 2001. Vol. 148 no. 1:1-7.
[5] J. Shentu and J. Armstrong. A new frequency offset estimator for OFDM. In Proc. Second Int. Symp. Commun. Syst. Networks Digital Signal Process. 2000. Vol. 32:13-16.
[6] B. G. Quinn. Estimating frequency by interpolation using Fourier coefficients. IEEE Trans. Signal Process. May 1994. Vol.42, no. 5:1264-1268,.
[7] B. Boashash, Estimating and interpreting the instantaneous frequency of a signal. I.Fundamentals, Proc. IEEE. Apr. 1992. Vol.80, no. 4:520-538.
[8] B. Boashash. Estimating and interpreting the instantaneous frequency of a signal. II.Algorithms and applications. Proc. IEEE. Apr. 1992. Vol.80, no. 4:540-568.
[9] Lopez-Salcedo, J. A. Vazquez, G. Frequency domain iterative pulse shape estimation based on second-order statistics. Signal Processing Advances in Wireless Communications,2004 IEEE 5th Workshop on 11-14 July 2004:92 - 96.
[10]Lei Huo, Tiandong Duan, Xiangqian Fang. A Novel Method of Modulation Classification for Digital Signals. International Joint Conference on Neural Networks.2006:2435-2438.
[11]Wei ZHANG, Hu YANG. Automatic Digital Modulation Classification Algorithms. ICSP 2006 Proceedings. 2006.
[12]Jiang Yuan, Zhang Zhao-Yang, Qiu Pei-Liang. Modulation Classification of Communication's Signals. MILCOM 2004 EEE Military Communications Conference.2004:1470-1478.
[13]Liang Hong and K. C. Ho, Classification of BPSK and QPSK signals with unknown signal level using the Bayes technique. Proc. IEEE ISCAS' 2003. May 2003, Vol.4, 1-4: 25-28.
[14]Tao HE, Xiaorong JING-. Modulation Classification Using ARBF Networks. ICSP'04 Proceedings. 2004:1809-1812.
[15]Jon Hamkins. Modulation Classification of MPSK for Space Applications. IEEE GLOBECOM 2006 proceedings.2006.
[16]Jeremy R. Waller and Gary D. Brushe. A Method For Differentiating Between Frequency And Phase Modulation Signals. Defense Science and Technology Organization.
[17]Eriko MATSUZAKI,Koichi ICHIGE,Hiroyuki AMI.An automatic recognition algorithm of analogue modulated signals for disturbance rejection.2003 IEEE Proceedings.2003:165-169.
[18]Dobre,A.Abdi,Y.Bar-Ness and W.Su.Survey of automatic modulation classification.IET Communication.2007.Vol.1.(2):137-156.
[19]Walter AKMOUCHE.Hierarchical digital modulation classification using cumulants.1999Proceeding of IEEE.1999:432-435.
[20]Ananthram Swami,Brian M.Sadler.Hierarchical Digital Modulation classification Using Cumulants.IEEE Transactions On Communications.March 2000.Vol.48,NO.3:416-429.
[21]Lei shen,Shiju Li,Chenseng song,etc.Automatic Modulation Classification of MPSK signals Using High Order Cumulants.ICSP 2006 Proceedings.2006:1356-1359.
[22]Luokun Liu,Jiadong Xu.A Novel Modulation Classification Method Based on High Order Cumulants.ICSP 2006 IEEE.2006:296-273.
[23]M.L.D.Wong,A.K.Nandi.Automatic digital modulation recognition using artificial neural network and genetic algorithm.Signal Processing 84(2004).2004:351-365.
[24]Zhao Yaqin,Ren Guanghui,Wang Xuexia.Automatic digital modulation recognition using artificial neural networks.IEEE Int.Conf.Neural Networks & Signal Processing.257-262
[25]孙仲康.单多基地有源无源定位技术.北京:国防工业出版社,2005.
[26]刘钰.无源定位技术研究及其定位精度分析:(硕士学位论文).西安:西北工业大学,2005.
[27]Clarkson P M.Optimal and adaptive signal processing.Florida:CRC Pr.,1993.
[28]Feintuch P L,Bershad N J,Reed F A.Time delay estimation using the LMS adaptive filter-dynamic behavior.IEEE Transactions on Signal Processing,1981,29(3):571-576.
[29]Ching P C,Chan Y T.Adaptive time delay estimation with constraints.IEEE Transactions on Acoustics,Speech,and Signal Processing,1988,36(4):599-602.
[30]王宏禹,邱天爽.自适应噪声抵消与时间延迟估计.大连:大连理工大学出版社,1999.
[31]Audoin B,Roux J.An innovative application of the Hilbert transform to time delay estimation of overlapped ultrasonic echoes.Ultrasonics in Elsevier Science,1996,34(1):25-33.
[32]Cabot R.A note on the application of the Hilbert transform to time delay estimation.IEEE Transactions on Signal Processing,1981,29(3):607-609.
[33]M.L.D.Wong,A.K.Nandi.Automatic digital modulation recognition using artificial neural network and genetic algorithm.Signal Processing 84(2004).2004:351-365.
[34]Zhao Yaqin,Ren Guanghui,Wang Xuexia.Automatic digital modulation recognition using artificial neural networks.IEEE Int.Conf.Neural Networks & Signal Processing.257-262
[35]J.Lopatka and M.Pedzisz.Automatic Modulation Classification Using Statistical Moments and a Fuzzy Classifier.IEEE.Proceeding of ICSP 2000.2000:1500-1506.
[36]Dimitar Kavalov,Victor Kalinin.Neural Network Surface Acoustic Wave RF Signal Processor for Digital Modulation Recognition.IEEE Transactions on Ultrasonics,Frerroelfctrics,and Frequency control.Sep.2002.Vol.49,No,9:1280-1292.
[37]胡延平,李广森,李纲等.利用参数统计方法白动识别数字调制信号.通信学报,2002,23(2):59-65.
[38]李杨,李国通,杨根庆.通信信号数字调制方式自动识别算法研究.电子与信息学报,2005.Vol.27,(02):197-202.
[39]Scarbrough K,Ahmed N,Carter G.On the simulation of a class of time delay estimation algorithms.IEEE Transactions on Acoustics,Speech,and Signal Processing,1981,29(3):534-540.
[40]Knapp C,Carter G.The generalized correlation method for estimation of time delay.IEEE Transactions on Acoustics,Speech,and Signal Processing,1976,24(4):320-327.
[41]Azaria M,Hertz D.Time delay estimation by generalized cross correlation methods.IEEE Transactions on Acoustics,Speech,and Signal Processing,1984,32(2):280-285.
[42]Miller L E,Lee J S.Error analysis of time delay estimation using a finite integration time correlator.IEEE Transactions on Acoustics,Speech,and Signal Processing,1981,29(3):490-496.
[43]Hertz D.Time delay estimation by combining efficient algorithms and generalized cross-correlation methods.IEEE Transactions on Acoustics,Speech,and Signal Processing,1986,34(1):1-7.
[44]Quazi A H.An overview on the time delay estimate in active and passive systems for target localization.IEEE Transactions on Acoustics,Speech,and Signal Processing,1981,29(3):527-533.
[45]邱天爽,汪琏.广义相关时间延迟估计的自适应实现.海洋技术,1994,13(4):20-31.
[46]Piersol A G.Time delay estimation using phase data.IEEE Transactions on Acoustics,Speech,and Signal Processing,1981,29(3):471-477.
[47]Youn D,Chiou S,Mathews V.Adaptive realization of the phase transform for time delay estimation.IEEE International Conference on Acoustics,Speech,and Signal Processing,New York,1984,9(1):648-651.
[48]Hinich M J,Wilson G R.Time delay estimation using the cross bispectrum.IEEE Transactions on Signal Processing,1992,40(1):106-113.
[49]Persson L,Asraf D E,Sigray Pet al.Bispectral analysis of underwater EM pulses.Proceedings of the IEEE signal processing workshop on higher-order statistics,Caesarea,1999:357-361.
[50]Carter G C.Coherence and time delay estimation.Proceedings of the IEEE,Los Alamitos,1987,75(2):236-255.
[2] O. Besson, M. Ghogho, and A. Swami. Parameter estimation for random amplitude chirp signals. IEEE Trans. Signal Processing. Dec. 1999. Vol. 47:3208-3219.
[3] M. Ghogho, A. Swami, and B. Garel. Performance analysis of cyclic statistics for the estimation of harmonics in multiplicative and additive noise. IEEE Trans. Signal Processing. Dec. 1999. Vol.47:3235-3249.
[4] Y. V. Zakharov, V. M. Baronkin, and T. C. Tozer. DFT-based frequency estimators with narrow acquisition range. Proc. Inst. Elect. Eng. Commun. 2001. Vol. 148 no. 1:1-7.
[5] J. Shentu and J. Armstrong. A new frequency offset estimator for OFDM. In Proc. Second Int. Symp. Commun. Syst. Networks Digital Signal Process. 2000. Vol. 32:13-16.
[6] B. G. Quinn. Estimating frequency by interpolation using Fourier coefficients. IEEE Trans. Signal Process. May 1994. Vol.42, no. 5:1264-1268,.
[7] B. Boashash, Estimating and interpreting the instantaneous frequency of a signal. I.Fundamentals, Proc. IEEE. Apr. 1992. Vol.80, no. 4:520-538.
[8] B. Boashash. Estimating and interpreting the instantaneous frequency of a signal. II.Algorithms and applications. Proc. IEEE. Apr. 1992. Vol.80, no. 4:540-568.
[9] Lopez-Salcedo, J. A. Vazquez, G. Frequency domain iterative pulse shape estimation based on second-order statistics. Signal Processing Advances in Wireless Communications,2004 IEEE 5th Workshop on 11-14 July 2004:92 - 96.
[10]Lei Huo, Tiandong Duan, Xiangqian Fang. A Novel Method of Modulation Classification for Digital Signals. International Joint Conference on Neural Networks.2006:2435-2438.
[11]Wei ZHANG, Hu YANG. Automatic Digital Modulation Classification Algorithms. ICSP 2006 Proceedings. 2006.
[12]Jiang Yuan, Zhang Zhao-Yang, Qiu Pei-Liang. Modulation Classification of Communication's Signals. MILCOM 2004 EEE Military Communications Conference.2004:1470-1478.
[13]Liang Hong and K. C. Ho, Classification of BPSK and QPSK signals with unknown signal level using the Bayes technique. Proc. IEEE ISCAS' 2003. May 2003, Vol.4, 1-4: 25-28.
[14]Tao HE, Xiaorong JING-. Modulation Classification Using ARBF Networks. ICSP'04 Proceedings. 2004:1809-1812.
[15]Jon Hamkins. Modulation Classification of MPSK for Space Applications. IEEE GLOBECOM 2006 proceedings.2006.
[16]Jeremy R. Waller and Gary D. Brushe. A Method For Differentiating Between Frequency And Phase Modulation Signals. Defense Science and Technology Organization.
[17]Eriko MATSUZAKI,Koichi ICHIGE,Hiroyuki AMI.An automatic recognition algorithm of analogue modulated signals for disturbance rejection.2003 IEEE Proceedings.2003:165-169.
[18]Dobre,A.Abdi,Y.Bar-Ness and W.Su.Survey of automatic modulation classification.IET Communication.2007.Vol.1.(2):137-156.
[19]Walter AKMOUCHE.Hierarchical digital modulation classification using cumulants.1999Proceeding of IEEE.1999:432-435.
[20]Ananthram Swami,Brian M.Sadler.Hierarchical Digital Modulation classification Using Cumulants.IEEE Transactions On Communications.March 2000.Vol.48,NO.3:416-429.
[21]Lei shen,Shiju Li,Chenseng song,etc.Automatic Modulation Classification of MPSK signals Using High Order Cumulants.ICSP 2006 Proceedings.2006:1356-1359.
[22]Luokun Liu,Jiadong Xu.A Novel Modulation Classification Method Based on High Order Cumulants.ICSP 2006 IEEE.2006:296-273.
[23]M.L.D.Wong,A.K.Nandi.Automatic digital modulation recognition using artificial neural network and genetic algorithm.Signal Processing 84(2004).2004:351-365.
[24]Zhao Yaqin,Ren Guanghui,Wang Xuexia.Automatic digital modulation recognition using artificial neural networks.IEEE Int.Conf.Neural Networks & Signal Processing.257-262
[25]孙仲康.单多基地有源无源定位技术.北京:国防工业出版社,2005.
[26]刘钰.无源定位技术研究及其定位精度分析:(硕士学位论文).西安:西北工业大学,2005.
[27]Clarkson P M.Optimal and adaptive signal processing.Florida:CRC Pr.,1993.
[28]Feintuch P L,Bershad N J,Reed F A.Time delay estimation using the LMS adaptive filter-dynamic behavior.IEEE Transactions on Signal Processing,1981,29(3):571-576.
[29]Ching P C,Chan Y T.Adaptive time delay estimation with constraints.IEEE Transactions on Acoustics,Speech,and Signal Processing,1988,36(4):599-602.
[30]王宏禹,邱天爽.自适应噪声抵消与时间延迟估计.大连:大连理工大学出版社,1999.
[31]Audoin B,Roux J.An innovative application of the Hilbert transform to time delay estimation of overlapped ultrasonic echoes.Ultrasonics in Elsevier Science,1996,34(1):25-33.
[32]Cabot R.A note on the application of the Hilbert transform to time delay estimation.IEEE Transactions on Signal Processing,1981,29(3):607-609.
[33]M.L.D.Wong,A.K.Nandi.Automatic digital modulation recognition using artificial neural network and genetic algorithm.Signal Processing 84(2004).2004:351-365.
[34]Zhao Yaqin,Ren Guanghui,Wang Xuexia.Automatic digital modulation recognition using artificial neural networks.IEEE Int.Conf.Neural Networks & Signal Processing.257-262
[35]J.Lopatka and M.Pedzisz.Automatic Modulation Classification Using Statistical Moments and a Fuzzy Classifier.IEEE.Proceeding of ICSP 2000.2000:1500-1506.
[36]Dimitar Kavalov,Victor Kalinin.Neural Network Surface Acoustic Wave RF Signal Processor for Digital Modulation Recognition.IEEE Transactions on Ultrasonics,Frerroelfctrics,and Frequency control.Sep.2002.Vol.49,No,9:1280-1292.
[37]胡延平,李广森,李纲等.利用参数统计方法白动识别数字调制信号.通信学报,2002,23(2):59-65.
[38]李杨,李国通,杨根庆.通信信号数字调制方式自动识别算法研究.电子与信息学报,2005.Vol.27,(02):197-202.
[39]Scarbrough K,Ahmed N,Carter G.On the simulation of a class of time delay estimation algorithms.IEEE Transactions on Acoustics,Speech,and Signal Processing,1981,29(3):534-540.
[40]Knapp C,Carter G.The generalized correlation method for estimation of time delay.IEEE Transactions on Acoustics,Speech,and Signal Processing,1976,24(4):320-327.
[41]Azaria M,Hertz D.Time delay estimation by generalized cross correlation methods.IEEE Transactions on Acoustics,Speech,and Signal Processing,1984,32(2):280-285.
[42]Miller L E,Lee J S.Error analysis of time delay estimation using a finite integration time correlator.IEEE Transactions on Acoustics,Speech,and Signal Processing,1981,29(3):490-496.
[43]Hertz D.Time delay estimation by combining efficient algorithms and generalized cross-correlation methods.IEEE Transactions on Acoustics,Speech,and Signal Processing,1986,34(1):1-7.
[44]Quazi A H.An overview on the time delay estimate in active and passive systems for target localization.IEEE Transactions on Acoustics,Speech,and Signal Processing,1981,29(3):527-533.
[45]邱天爽,汪琏.广义相关时间延迟估计的自适应实现.海洋技术,1994,13(4):20-31.
[46]Piersol A G.Time delay estimation using phase data.IEEE Transactions on Acoustics,Speech,and Signal Processing,1981,29(3):471-477.
[47]Youn D,Chiou S,Mathews V.Adaptive realization of the phase transform for time delay estimation.IEEE International Conference on Acoustics,Speech,and Signal Processing,New York,1984,9(1):648-651.
[48]Hinich M J,Wilson G R.Time delay estimation using the cross bispectrum.IEEE Transactions on Signal Processing,1992,40(1):106-113.
[49]Persson L,Asraf D E,Sigray Pet al.Bispectral analysis of underwater EM pulses.Proceedings of the IEEE signal processing workshop on higher-order statistics,Caesarea,1999:357-361.
[50]Carter G C.Coherence and time delay estimation.Proceedings of the IEEE,Los Alamitos,1987,75(2):236-255.