Title :
Power Spectrum Estimation by using Digital Frequency Tracking Filter
Author_Institution :
Electrical Engineering Department, California State University, Long Beach, Long Beach, CA 90840
Abstract :
In this paper, three approaches of estimating the spectrum by using digital frequency tracking filter are presented. The common structure of these approaches are. (1) to have a bank of filters to process the received signals (signals are seperated into a few sections as the excitation frequency sweeps from low to high), (2) to use the Welsh method to compute the spectral density via the FFT (Fast Fourier Transform) algorithm. The bandpass filter bank is employed in the 1st approach to remove noises which are contained in all sections of received signals. The decimation and interpolation schemes are used is the 2nd approach to reduce the unnecessary computations which exist in the over sampled sections of received signals. The modulation and demodulation schemes (for high-frequency sections only) combined with decimation and interpolation schemes (for over sampled section only) are employed in the 3rd approach.
Keywords :
Band pass filters; Costs; Demodulation; Design engineering; Digital filters; Filter bank; Frequency estimation; Interpolation; Spectral analysis; Testing;
Conference_Titel :
American Control Conference, 1986
Conference_Location :
Seattle, WA, USA