Title :
Polynomial constrained LMS adaptive algorithm for measurement signal processing
Author_Institution :
Dept. of Inf. Technol., Tampere Univ. of Technol., Finland
Abstract :
The LMS adaptive algorithm is modified in such a way that unbiased polynomial estimation or prediction is guaranteed continuously. This is accomplished by introducing the necessary linear constraints and modifying the coefficient update equations accordingly. The adaptive properties in polynomial prediction allow optimum noise attenuation even when the characteristics of noise and interference are dynamically changing. Also, adaptive differentiation with the constrained LMS algorithm is proposed. Examples of signal processing in a velocity measurement application are shown.
Keywords :
differentiation; least mean squares methods; measurement systems; polynomials; signal processing; velocity measurement; LMS adaptive algorithm; adaptive properties; coefficient update equations; interference; measurement signal processing; noise; optimum noise attenuation; polynomial constrained LMS adaptive algorithm; polynomial prediction; unbiased polynomial estimation; unbiased polynomial prediction; velocity measurement; Adaptive algorithm; Adaptive signal processing; Attenuation; Extrapolation; Finite impulse response filter; Interference constraints; Least squares approximation; Polynomials; Signal processing algorithms; Smoothing methods;
Conference_Titel :
IECON 02 [Industrial Electronics Society, IEEE 2002 28th Annual Conference of the]
Print_ISBN :
0-7803-7474-6
DOI :
10.1109/IECON.2002.1185497
