Abstract

Time aliasing methods of spectrum estimation alter the time representation of a signal for the purpose of improving its frequency domain representation. Time aliasing allows characteristics of longer time window functions to be used with shorter DFT length. Windows designed specifically for use with time aliasing have improved properties compared to conventional windows. Many previous uses of time aliasing, including overlap-and-add methods of digital filtering, have focused on the elimination of time aliasing effects in the frequency domain in order to improve the representation of reconstructed signals in the time domain and have not addressed the issues associated with spectrum analysis. Proper use of time aliasing methods of spectrum analysis requires understanding of time and frequency scaling effects resulting from using a longer effective time window with a given DFT length and the effects of spectral averaging on the time aliased spectral estimate. Time aliasing has been shown to reduce bias error in spectral estimates by reducing spectral leakage and improving effective frequency resolution, particularly in regions of high dynamic range in the spectrum, yielding improved measurements of spectra containing narrowband phenomena such as those encountered in the applications of system identification and rotating machinery signature analysis.

Degree

PhD

College and Department

Ira A. Fulton College of Engineering and Technology; Mechanical Engineering

Rights

http://lib.byu.edu/about/copyright/