ESTIMATION OF ACOUSTIC SOURCE STRENGTH BY INVERSE METHODS: PART II, EXPERIMENTAL INVESTIGATION OF METHODS FOR CHOOSING REGULARIZATION PARAMETERS

Two regularization methods, Tikhonov regularization and singular value discarding, are used to improve the accuracy of reconstruction of acoustic source strength by inverse techniques. In this paper, some methods are investigated for choosing the Tikhonov regularization parameter and the singular values to be discarded. Of these, we concentrate on the use of ordinary cross-validation and generalized cross-validation. These methods can provide an appropriate regularization parameter without prior knowledge of either the acoustic source strength or the contaminating measurement noise. Some experimental results obtained using a randomly vibrating simply supported plate mounted in a baffle are presented to illustrate the performance of the methods for choosing the regularization parameters.

[1]  Peter Craven,et al.  Smoothing noisy data with spline functions , 1978 .

[2]  Dianne P. O'Leary,et al.  The Use of the L-Curve in the Regularization of Discrete Ill-Posed Problems , 1993, SIAM J. Sci. Comput..

[3]  D. Titterington,et al.  A cautionary note about crossvalidatory choice , 1989 .

[4]  David M. Allen,et al.  The Relationship Between Variable Selection and Data Agumentation and a Method for Prediction , 1974 .

[5]  W. Seering,et al.  Multichannel Structural Inverse Filtering , 1984 .

[6]  N. A. Halliwell,et al.  Laser-Doppler measurement of vibrating surfaces: A portable instrument , 1979 .

[7]  Nikolas P. Galatsanos,et al.  Methods for choosing the regularization parameter and estimating the noise variance in image restoration and their relation , 1992, IEEE Trans. Image Process..

[8]  E. Rothwell,et al.  A unified approach to solving ill‐conditioned matrix problems , 1989 .

[9]  P. Nelson,et al.  ESTIMATION OF ACOUSTIC SOURCE STRENGTH BY INVERSE METHODS: PART I, CONDITIONING OF THE INVERSE PROBLEM , 2000 .

[10]  P. Bloomfield,et al.  Numerical differentiation procedures for non-exact data , 1974 .

[11]  D. M. Titterington,et al.  A Study of Methods of Choosing the Smoothing Parameter in Image Restoration by Regularization , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  R. Blevins,et al.  Formulas for natural frequency and mode shape , 1984 .

[13]  A. Zinober.  Matrices: Methods and Applications , 1992 .

[14]  G. Dunteman.  Principal Components Analysis , 1989 .

[15]  Frank Fahy,et al.  Application of an area-integrating vibration velocity transducer , 1996 .

[16]  W J Krzanowski,et al.  Cross-Validation for Choosing the Number of Important Components in Principal Component Analysis. , 1995, Multivariate behavioral research.