Title :
Optimization of generalized predictors
Author :
Koppinen, Konsta ; Yli-Hietanen, Jari ; Astola, Jaakko
Author_Institution :
Signal Process. Lab., Tampere Univ. of Technol., Finland
Abstract :
Polynomial FIR predictors are generalized to include also the prediction of sinusoidal signals. A general optimization method is presented for optimizing these predictors, which gives optimal FIR predictors with respect to the noise gain, the autocorrelation matrix of additive noise and frequency response as special cases
Keywords :
FIR filters; approximation theory; filtering theory; frequency response; matrix algebra; optimisation; polynomials; prediction theory; random noise; signal reconstruction; additive noise; autocorrelation matrix; frequency response; noise gain; optimal FIR predictors; optimization; polynomial FIR predictors; prediction of sinusoidal signals; Additive noise; Design methodology; Finite impulse response filter; Laboratories; Nonlinear filters; Optimization methods; Polynomials; Predictive models; Signal processing; Time domain analysis;
Conference_Titel :
Instrumentation and Measurement Technology Conference, 1997. IMTC/97. Proceedings. Sensing, Processing, Networking., IEEE
Conference_Location :
Ottawa, Ont.
Print_ISBN :
0-7803-3747-6
DOI :
10.1109/IMTC.1997.603916
