Title :
Signal estimation using adapted tree-structured bases and the MDL principle
Author_Institution :
Beckman Inst. for Adv. Sci. & Technol., Illinois Univ., Urbana, IL, USA
Abstract :
We present a new signal denoising technique based on adapted tree-structured bases and the following paradigms: (a) use of the minimum description length (MDL) principle for estimating the signal in each branch of the tree and the best tree structure; (b) use of signal-adapted filter banks
Keywords :
adaptive filters; adaptive signal processing; band-pass filters; filtering theory; noise; parameter estimation; signal representation; trees (mathematics); MDL; adapted tree-structured bases; minimum description length; optimal tree design; signal denoising technique; signal estimation; signal representation; signal-adapted filter banks; tree structure; Cost function; Estimation; Filter bank; Minimax techniques; Signal denoising; Testing; Tree data structures; Wavelet packets;
Conference_Titel :
Time-Frequency and Time-Scale Analysis, 1996., Proceedings of the IEEE-SP International Symposium on
Conference_Location :
Paris
Print_ISBN :
0-7803-3512-0
DOI :
10.1109/TFSA.1996.546706
