p.246
p.251
p.259
p.266
p.273
p.283
p.288
p.293
p.303
Spectral Error Analysis of Windowed Discrete Sinusoidal Signals
Abstract:
This paper presents an analysis of spectral error analysis of windowed discrete sinusoidal signals. The objective of the work is to understand how window functions affect the discrete frequency spectrum and how to minimize spectrum errors. The spectral error analysis, in which the mathematical expression of the spectral error is derived, is conducted to a discrete sinusoidal signal with single frequency. The analysis reveals the fact that the applied window functions will produce peak spectral error regardless of the sampling rate and recording time. This result has been confirmed through numerical simulations. In addition, an experiment is conducted to investigate another type of spectral error, which is frequency error. The frequency error can be minimized by collecting more samples in signal measurement process, which means the recording time is increased, even though this effort does not improve the amplitude accuracy.
Info:
Periodical:
Pages:
273-279
Citation:
Online since:
June 2016
Authors:
Price:
Сopyright:
© 2016 Trans Tech Publications Ltd. All Rights Reserved
Share:
Citation: