For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Parallel Implementation of a Kalman-Based Sinusoidal Estimator
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2001/12/01
Print ISSN: 0916-8508
Type of Manuscript: LETTER
Category: Digital Signal Processing
nonlinear filter, Kalman filter, frequency estimation, filter bank, decimation, sinusoid,
Full Text: PDF>>
Phase-based methods for estimating the frequency of a sinusoid have typically suffered from a threshold effect, where for signal to noise ratio (SNR) below the threshold, the mean squared error of the estimate rapidly increases. Furthermore, it is a significant problem that the threshold is considerably high and strongly depends on frequency. To overcome the difficulties, a Kalman-based sinusoidal estimator bank (KSEB) is proposed. In the derivation of the KSEB, a four-channel filter bank and decimation technique are effectively used. The computer simulation also demonstrates the superiority of the KSEB to the other frequency estimators.