ELIMINATION OF THE MEASUREMENT NOISE BY USING WAVELET TRANSFORMATIONS

This contribution describes a method for ideal de-noising. This method is based on a choice of an optimal basis from a library of wavelet orthogonal bases. This is the process of minimizing of the cost functional of entropy in contribution with shrinkage method. In particular, the chosen shrinkage methods are the estimation of standard deviation σ and of Donoho-Johnstone universal threshold λ for the wavelet coefficients. The chosen entropy is Shannon entropy and is defined by the discrete distribution of squared wavelet coefficients. The Donoho thresholding rule belongs to the simplest non-linear shrinkage technique. The result of this ideal de-noising is presented on one example. The algorithm, selecting an optimal basis from a library of wavelet bases for ideal de-noising, was created with using MATLAB and The Wavelet Toolbox