Title :
Adaptive noise cancelling with systolic arrays
Author :
Kshonze, Kristopher ; Steenaart, Willem
Author_Institution :
Dept. of Electr. Eng., Ottawa Univ., Ont., Canada
Abstract :
The QR decomposition with systolic array realization is compared to the LMS algorithm. The LMS algorithm proves to be less complex, very efficient under finite precision, and compatible in terms of speed (it can achieve the same speed at signal-to-noise ratio expense). In terms of misadjustment the LMS always produces self noise, whereas the QR decomposition with systolic array can give an optimum solution.<>
Keywords :
adaptive filters; cellular arrays; interference suppression; LMS algorithm; QR decomposition; optimum solution; self noise; signal-to-noise ratio; systolic arrays; Adaptive filters; Eigenvalues and eigenfunctions; Equations; Least squares approximation; Least squares methods; Matrix decomposition; Noise cancellation; Open loop systems; Signal to noise ratio; Systolic arrays;
Conference_Titel :
Circuits and Systems, 1988., IEEE International Symposium on
Conference_Location :
Espoo, Finland
DOI :
10.1109/ISCAS.1988.14977
