Title :
The use of the modified escalator algorithm to improve the performance of transform-domain LMS adaptive filters
Author :
Parikh, Vipul N. ; Baraniecki, Anna Z.
Author_Institution :
Voxware Inc., Princeton, NJ., USA
fDate :
3/1/1998 12:00:00 AM
Abstract :
This paper describes a new algorithm that improves the convergence performance of the transform-domain least mean-square (TRLMS) algorithm. The algorithm exploits the sparse structure of the correlation matrix of the transformed input process to derive a data dependent Gram-Schmidt orthogonalization type transform of the process. We show its faster convergence compared with the time-domain least mean-square (LMS) algorithm and the DCT or the DWT-based TRLMS algorithm. The Gram-Schmidt orthogonalization is realized using a modified adaptive escalator algorithm. The modification significantly reduces the computational complexity of the adaptive escalator algorithm and determines the computational complexity of the proposed algorithm
Keywords :
adaptive filters; adaptive signal processing; computational complexity; convergence of numerical methods; correlation methods; filtering theory; least mean squares methods; sparse matrices; transforms; DCT; DWT-based TRLMS algorithm; Gram-Schmidt orthogonalization type transform; LMS algorithm; TRLMS algorithm; computational complexity reduction; convergence performance; correlation matrix; data dependent transform; modified adaptive escalator algorithm; sparse structure; time-domain least mean-square; transform-domain LMS adaptive filters; transformed input process; Computational complexity; Convergence; Discrete cosine transforms; Discrete transforms; Discrete wavelet transforms; Karhunen-Loeve transforms; Least squares approximation; Signal processing; Signal processing algorithms; Sparse matrices;
Journal_Title :
Signal Processing, IEEE Transactions on
