The Hyperbolic Class of Quadratic Time-Frequency Representations-Part II: Subclasses, Intersection w - Signal Processing, IEEE Transactions on
نویسندگان
چکیده
Part I of this paper introduced the hyperbolic class (HC) of quadratic/bilinear time-frequency representations (QTFR’s) as a new framework for constant-Q time-frequency analysis. The present Part II defines and studies the following four subclasses of the HC: • The localized-kernel subclass of the HC is related to a timefrequency concentration property of QTFR’s. It is analogous to the localized-kernel subclass of the affine QTFR class. • The affine subclass of the HC (affine HC) consists of all HC QTFR’s that satisfy the conventional time-shift covariance property. It forms the intersection of the HC with the affine QTFR class. • The power subclasses of the HC consist of all HC QTFR’s that satisfy a “power time-shift” covariance property. They form the intersection of the HC with the recently introduced power classes. • The power-warp subclass of the HC consists of all HC QTFR’s that satisfy a covariance to power-law frequency warpings. It is the HC counterpart of the shift-scale covariant subclass of Cohen’s class. All of these subclasses are characterized by 1-D kernel functions. It is shown that the affine HC is contained in both the localizedkernel hyperbolic subclass and the localized-kernel affine subclass and that any affine HC QTFR can be derived from the Bertrand unitary P0-distribution by a convolution. We furthermore consider the properties of regularity (invertibility of a QTFR) and unitarity (preservation of inner products, Moyal’s formula) in the HC. The calculus of inverse kernels is developed, and important implications of regularity and unitarity are summarized. The results comprise a general method for least-squares signal synthesis and new relations for the AltesMarinovich Q-distribution.
منابع مشابه
The hyperbolic class of quadratic time-frequency representations. II. Subclasses, intersection with the affine and power classes, regularity, and unitarity
Part I of this paper introduced the hyperbolic class (HC) of quadratic/bilinear time-frequency representations (QTFR’s) as a new framework for constant-Q time-frequency analysis. The present Part II defines and studies the following four subclasses of the HC: • The localized-kernel subclass of the HC is related to a timefrequency concentration property of QTFR’s. It is analogous to the localize...
متن کاملThe hyperbolic class of quadratic time-frequency representations. I. Constant-Q warping, the hyperbolic paradigm, properties, and members
The time-frequency (TF) version of the wavelet transform and the “affine” quadraticlbilinear TF representations can be used for a TF analysis with constant-Q characteristic. This paper considers a new approach to constant-Q TF analysis. A specific TF warping transform is applied to Cohen’s class of quadratic TF representations, which results in a new class of quadratic TF representations with c...
متن کاملOptimal quadratic detection and estimation using generalized joint signal representations
Time-frequency analysis has recently undergone signiicant advances in two main directions: statistically optimized methods that extend the scope of time-frequency-based techniques from merely exploratory data analysis to more quantitative application, and generalized joint signal representations that extend time-frequency-based methods to a richer class of nonstationary signals. This paper fuse...
متن کاملThe power classes-quadratic time-frequency representations with scale covariance and dispersive time-shift covariance
We consider scale-covariant quadratic time– frequency representations (QTFR’s) specifically suited for the analysis of signals passing through dispersive systems. These QTFR’s satisfy a scale covariance property that is equal to the scale covariance property satisfied by the continuous wavelet transform and a covariance property with respect to generalized time shifts. We derive an existence/re...
متن کاملOptimizing time-frequency kernels for classification
In many pattern recognition applications, features are traditionally extracted from standard time–frequency representations (TFRs). This assumes that the implicit smoothing of, say, a spectrogram is appropriate for the classification task. Making such assumptions may degrade classification performance. In general, any time–frequency classification technique that uses a singular quadratic TFR (e...
متن کامل