Abstract
This paper addresses time-frequency (TF) analysis from a statistical signal processing perspective, with the goal of developing a general statistical methodology for TF analysis. We review earlier work on statistical models for nonstationary stochastic signals, including frequency modulated locally stationary processes, which have covariance functions which yield nonnegative Wigner distributions. For such processes, time- frequency spectra may be defined without invoking 'local-' or 'quasi-' stationarity. These results are extended to include general time-varying linear systems and their associated time- frequency spectra. Any time-varying linear system driven by white noise has associated with it a nonnegative time-frequency spectrum. The bilinear class of time-frequency distributions are estimators of this time-frequency spectrum, as are adaptive methods such as positive time-frequency distributions and adaptive multitaper spectrograms. An analysis of the statistical properties of these estimators, including moments and distributional properties, is reviewed.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
