本論文內容以多解析盲蔽反捲積創新演算法為主，此演算法是以小波轉換中Mallat分解及重構演算法為基礎，將訊號正交分解成不同解析度子空間的特點，以倒雙譜演算法計算各子空間訊號系統函數，恢復各子空間訊號，最後再將各子空間訊號重構為與原輸入訊號相同解析度之時域訊號輸出；以多解析盲蔽反捲積演算法應用於南化水庫荷本閥自然頻率和其對應之阻尼比計算，及在南海淺海調頻訊號時間延遲估計兩計畫為例，說明觀測訊號在低訊雜比的情況下，使用多解析盲蔽反捲積演算法，可提高訊雜比、恢復觀測訊號、及增加訊號辨識能力；另對於旋轉機械故障訊號診斷中，比較傳統以二階統計分析為基礎之頻譜與三階統計分析為基礎之雙譜，在旋轉機械振動訊號中，提取故障特徵方式同時進行比較及討論。

　　In this paper, a so-called “the multiresolution blind deconvolution algorithm”(MBD) is presented. The signal is to decompose a multicomponent signal into a number of single component signals by using Mallat algorithm. MBD algorithm with biceptrum method is used to reconstruct the single component signal. Then, the enhanced signal is obtained after reconstructing and combining each single component. The new process is characterized by its strong robustness against additive Gaussian noise.
　　Three representative acoustical problems are presented in this paper. First, the application of MBD algorithm is illustrated. Two experimental results, the computation of natural frequency and its damping ratio of the Howell-Bunger gate located at the bottom of a dam and arrival time delay estimation in South China Sea, are presented. Then the comparison of spectrum and bispectrum applied in condition monitoring of rotating machine is discussed.

[1]. P. J. Tavner and J. Penman, “Condition monitoring of electrical machines”, Letchworth”, U.K. Res. Studies, 1987.
[2]. T. W. Cooley, J. W. Tukey, “An algorithem for the machine calculation of complex fourier series”. Mathematics of Computation, 19, 297-301, 1965.
[3]. C. C. Wang, G. P.James Too, “Rotating machine fault detection based on HOS and artificial neural networks”, J. of Intelligent Manufacturing, 13(4), 283-294. 2002.
[4]. I. Daubechies, Ten Lectures On Wavelets. SIAM: 258-259, 1992.
[5]. D. Lee, “Condition monitoring of electrical equipment using wavelet analysis”, 1998 IEEE International Conference on Systems, Man, and Cybernetics, 1, 811 – 816, 1998.
[6]. X. Li, S. K. Tso, J. Wang, “Real-time tool condition monitoring using wavelet transforms and fuzzy techniques”, IEEE Transactions on Systems, Man and Cybernetics, Part C, 30, 3, 2000.
[7]. T. Kaewkongka, “Continuous wavelet transform and neural network for condition monitoring of rotodynamic machinery”, Instrumentation and Measurement Technology Conference, Proceedings of the 18th IEEE, 3, 962 – 966,2001.
[8]. M. R. Raghuveer, “Time domain approaches to quadratic phase coupling estimation”, IEEE Trans. on Automatic Control, 35, 48-56, 1990.
[9]. W. Q. Jeffries, J. A. Chambers, D. G. Infield, “Experience with bicoherence of electrical power for condition monitoring of wind turbine blades”, IEE Proceedings, Image and Signal Processing, , 145(3) , 141 – 148, 1998.
[10]. I. Lasurt, A. F. Stronach, J. Penman, “A fuzzy logic approach to the interpretation of higher order spectra applied to fault diagnosis in electrical machines”, 19th International Conference of the North American, Fuzzy Information Processing Society, 13-15, 158-162, 2000
[11]. A. C. McCormick, A. K. Nandi, “Bispectral and trispectral features for machine conditiondiagnosis”, IEE Proceedings, Image and Signal processing, 146(5), 229-234. 1999.
[12]. R. D. Priebe, G. R. Wilson, “Wavelet applications to structural analysis Acoustics, Speech, and Signal Processing”, 1994 IEEE International Conference on Acoustics, Speech, Signal Processing, 3, 205-208, 1994.
[13]. E. P. Wigner, “On the quantum correction for thermodynamic equilibrium”, Phys. Rev., 40, 749-759, 1932.
[14]. J. Vill, “Theorie et applications de la notion de signal analytique”, Cable et Transmission, 2A, 61-74, 1948.
[15]. S. Barbarossa, “Analysis of multicomponent LFM signals by a combined Wigner-Hough transform”, IEEE Trans. Signal Proc., 143, 1511-1515, 1995.
[16]. S. Haykin, Adaptive filter theory, Prentice-Hall, 1999.
[17]. D. S. Luo, A. E. Yagle, “Lattice algorithms applied to the blind deconvolution problem”, International Conference on Speech, and Signal Processing, 2341 – 2344, 1989.
[18]. W. P. Tai, “Blind deconvolution by self-organization”, International Conference on Neural Networks, 1568 – 1573, 1997.
[19]. A. D. Back, “Blind deconvolution of signals using a complex recurrent network”, 1994 IEEE Workshop, Neural Networks for Signal Processing, 5, 565 – 574, 1994.
[20]. M. S. Bartlett, An introduction to stochastics processes, UK, Cambridge University Press, 1955.
[21]. D. R. Brillinger, M. Roseblatt, Compution and interpretation of kth-order spectra, in Spectral Analysis of time Series, Wiley, 189-232, 1967.
[22]. C. L. Nikias, J. M.Mendel. “Signal processing with higher-order spectra”, IEEE Signal Processing Mag., 10(3), 10-37, 1993.
[23]. C. L. Nikias, R Pan, “ARMA modelling of fourth-order cumulants and phase estimation”, IEEE Trans. Circuits, Systems and Signal Processing, 7, 291-325. 1988.
[24]. C. L. Nikias, R. Pan, “Time delay estimation in unknown Gaussian spatially correlated noise”, IEEE Trans. Acoustics, Speech, Signal Processing, 36, 291-325, 1988.
[25]. C. L. Nikias, A. Petropulu, Higher-order spectra analysis: a nonlinear signal processing framework, Prentice-Hall, 1993.
[26]. A. Swami, “Third order Wigner distributions”, 1991 IEEE International Conference on Acoustics, Speech, Signal Processing, 3081-3084, 1991.
[27]. A. Swami, “Higher order Wigner distributions”, 1992 IEEE International Conference on Acoustics, Speech, Signal Processing,, 290-301, 1992.
[28]. M. Boumahdi, “Blind identification using the kurtosis with applications to field data”, Signal Processing, 48, 205-216, 1996.
[29]. J. K. Martin, A. K. Naadi, “Blind system identification using second, third and fourth order cumulants”, J. of the Franklin Institute of Science, 333, 1-13,1996.
[30]. T. Clapp, S. Godsill, “Bayesian blind deconvolution for mobile communications”, IEE Colloquium on Adaptive Signal Processing for Mobile Communication Systems, 29, 1-6, 1997.
[31]. C. Prati, F. Rocca, Y. Kost, E. Damonti, “Some results on blind deconvolution applied to digital communication signals”, 13th International Conference on DSP_ 97, 1 , 2-4, 1997
[32]. R. Sensing, “Blind deconvolution for doppler centroid estimation in high frequency SAR”, IEEE Transactions on Geoscience and Remote Sensing, 29(6), 934-941, 1991.
[33]. M. K. Broadhead,, L. A. Pflug, “Deconvolution for transient classification using fourth order statistics”, OCEANS '97. MTS/IEEE Conference Proceedings, 1, 436-440, 1997.
[34]. D. A. Caughey, R. L. Kirlin, “Blind deconvolution of echosounder envelopes”, IEEE International Conference on Acoustics, Speech, and Signal Processing, 6, 3149-3152, 1996.
[35]. I. Doroslovacki, “Wavelet-based linear system modeling and adaptive filtering”,IEEE Trans., Signal Proc., 44(5), 1156-1167, 1996.
[36]. G. E. Box, Time series analysis-forecasting and control, San Francisko, Hololden-Day, 1970.
[37]. G. B. Giannakis, “Cumulants : A powerful tool in signal processing”, IEEE Trans. Signal Proceedings, 75, 1333-1334. 1987.
[38]. C. L. Nikias, R Pan, “ARMA modelling of fourth-order cumulants and phase estimation”, Systems and Signal Processing, 7, 291-325, 1988.
[39]. A. P. Petropulu, C L. Nikias, “The complex cepstrum and bicepstrum: Analytic performance evaluation in the presence of Gaussian noise”, IEEE Trans. Acoustics, Speech, Signal Processing, 38, 1246-1256, 1990.
[40]. A. Petropulu, C. L. Nikias, “Blind deconvolution using signal reconstruction from partial higher-order cepstral information”, IEEE Trans. Signal Processing, 41, 2088-2095, 1993.
[41]. A. Croisier, D. Esteban, C. Galand, “Perfect channel splitting by use of interpolation/decimation/tree decomposition techniques”, in int. conf. on info on Patra, Greece, Sciences and Systems, 443-446, 1976.
[42]. S. Mallat, “A theory for multiresolution signal decomposition : the wavelet representation”, IEEE Trans. Patt. Recog. and Mach. Intell., 11(7), 674-693, 1989.
[43]. A. V. Oppenheim, Discrete-time signal processing, Prentice-Hall, 1998.