Title :
Soft thresholding by noise invalidation
Author :
Beheshti, Soosan ; Nikvand, Nima ; Fernando, Xavier N.
Author_Institution :
Dept. of Electr. & Comput. Eng., Ryerson Univ., Toronto, ON
Abstract :
A new thresholding technique for data denoising is proposed. Using statistical properties of additive noise, this method provides an adaptive data dependent soft threshold to remove the effects of noise. The observed data can be denoised in any orthogonal basis. The simulations demonstrate the advantages of the new approach for data denoising with wavelet transformation.
Keywords :
signal denoising; statistical analysis; wavelet transforms; adaptive data dependent soft threshold; additive noise; data denoising; noise invalidation; orthogonal basis; soft thresholding; statistical property; wavelet transformation; Additive noise; Gaussian noise; Noise reduction; Random processes; Statistics;
Conference_Titel :
Communications, 2008 24th Biennial Symposium on
Conference_Location :
Kingston, ON
Print_ISBN :
978-1-4244-1945-6
Electronic_ISBN :
978-1-4244-1946-3
DOI :
10.1109/BSC.2008.4563246
