Matching wavelet packets to Gaussian random processes

We consider the problem of approximating a set of arbitrary, discrete-time, Gaussian random processes by a single, representative wavelet-based, Gaussian process. We measure the similarity between the original processes and the wavelet-based process with the Bhattacharyya (1943) coefficient. By manipulating the Bhattacharyya coefficient, we reduce the task of defining the representative process to finding an optimal unitary matrix of wavelet-based eigenvectors, an associated diagonal matrix of eigenvalues, and a mean vector. The matching algorithm we derive maximizes the nonadditive Bhattacharyya coefficient in three steps: a migration algorithm that determines the best basis by searching through a wavelet packet tree for the optimal unitary matrix of wavelet-based eigenvectors; and two separate fixed-point algorithms that derive an appropriate set of eigenvalues and a mean vector. We illustrate the method with two different classes of processes: first-order Markov and bandlimited. The technique is also applied to the problem of robust terrain classification in polarimetric SAR images.

[1]  JoBea Way,et al.  Mapping of forest types in Alaskan boreal forests using SAR imagery , 1994, IEEE Trans. Geosci. Remote. Sens..

[2]  Gregory W. Wornell,et al.  A Karhunen-Loève-like expansion for 1/f processes via wavelets , 1990, IEEE Trans. Inf. Theory.

[3]  Carl Taswell.  Near-best basis selection algorithms with non-additive information cost functions , 1994, Proceedings of IEEE-SP International Symposium on Time- Frequency and Time-Scale Analysis.

[4]  N. Keshava,et al.  Wavelets and random processes: optimal matching in the Bhattacharyya sense , 1996, Conference Record of The Thirtieth Asilomar Conference on Signals, Systems and Computers.

[5]  Martin Vetterli,et al.  Orthogonal time-varying filter banks and wavelet packets , 1994, IEEE Trans. Signal Process..

[6]  Harry L. Van Trees,et al.  Detection, Estimation, and Modulation Theory, Part I , 1968 .

[7]  K Ramchandran,et al.  Best wavelet packet bases in a rate-distortion sense , 1993, IEEE Trans. Image Process..

[8]  Ronald R. Coifman,et al.  Entropy-based algorithms for best basis selection , 1992, IEEE Trans. Inf. Theory.

[9]  J. Douglas Faires,et al.  Numerical Analysis , 1981 .

[10]  Ravi Mazumdar,et al.  Wavelet representations of stochastic processes and multiresolution stochastic models , 1994, IEEE Trans. Signal Process..

[11]  Keinosuke Fukunaga,et al.  Introduction to Statistical Pattern Recognition , 1972 .

[12]  T. Kailath.  The Divergence and Bhattacharyya Distance Measures in Signal Selection , 1967 .

[13]  J. Miller.  Numerical Analysis , 1966, Nature.

[14]  José M. F. Moura,et al.  Terrain classification in polarimetric SAR using wavelet packets , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[15]  I. Daubechies.  Orthonormal bases of compactly supported wavelets , 1988 .

[16]  Michèle Basseville,et al.  Modeling and estimation of multiresolution stochastic processes , 1992, IEEE Trans. Inf. Theory.

[17]  Jun Zhang,et al.  A wavelet-based KL-like expansion for wide-sense stationary random processes , 1994, IEEE Trans. Signal Process..

[18]  Jose M. F. Moura,et al.  Robust classification of targets in POL-SAR using wavelet packets , 1997, Proceedings of the 1997 IEEE National Radar Conference.

[19]  Kannan Ramchandran,et al.  Tilings of the time-frequency plane: construction of arbitrary orthogonal bases and fast tiling algorithms , 1993, IEEE Trans. Signal Process..