Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
564310 | Digital Signal Processing | 2016 | 16 Pages |
Abstract
Results of an investigation of the characteristic estimator properties for periodically correlated time series obtained on the basis of finite data length are given. The formulae for the bias and variance of the estimators for mean and covariance function Fourier coefficients are found. The conditions for the choice of sampling interval value, for which aliasing effects do not appear, are obtained. The interpolation formulae for the mean and covariance function estimates are derived. The dependencies of the statistical characteristics of the estimators on sampling interval and sample size for modulated signals are analyzed.
Related Topics
Physical Sciences and Engineering
Computer Science
Signal Processing
Authors
Ihor Javors'kyj, Ivan Matsko, Roman Yuzefovych, Zbigniew Zakrzewski,