Showing papers on "Time–frequency analysis published in 2007"
TL;DR: The intrinsic time-scale decomposition (ITD) as discussed by the authors decomposes a signal into a sum of proper rotation components, for which instantaneous frequency and amplitude are well defined, and a monotonic trend.
Abstract: We introduce a new algorithm, the intrinsic time-scale decomposition (ITD), for efficient and precise time–frequency–energy (TFE) analysis of signals. The ITD method overcomes many of the limitations of both classical (e.g. Fourier transform or wavelet transform based) and more recent (empirical mode decomposition based) approaches to TFE analysis of signals that are nonlinear and/or non-stationary in nature. The ITD method decomposes a signal into (i) a sum of proper rotation components, for which instantaneous frequency and amplitude are well defined, and (ii) a monotonic trend. The decomposition preserves precise temporal information regarding signal critical points and riding waves, with a temporal resolution equal to the time-scale of extrema occurrence in the input signal. We also demonstrate how the ITD enables application of single-wave analysis and how this, in turn, leads to a powerful new class of real-time signal filters, which extract and utilize the inherent instantaneous amplitude and frequency/phase information in combination with other relevant morphological features.
329 citations
TL;DR: A technique to improve the fault detection technique by using the classical multiple signal classification (MUSIC) method and has been applied to detect a rotor broken bar fault in a three-phase squirrel-cage induction machine under different loads and in steady-state condition.
Abstract: Fault detection in alternating-current electrical machines that is based on frequency analysis of stator current has been the interest of many researchers. Several frequency estimation techniques have been developed and are used to help the induction machine fault detection and diagnosis. This paper presents a technique to improve the fault detection technique by using the classical multiple signal classification (MUSIC) method. This method is a powerful tool that extracts meaningful frequencies from the signal, and it has been widely used in different areas, which include electrical machines. In the proposed application, the fault sensitive frequencies have to be found in the stator current signature. They are numerous in a given frequency range, and they are affected by the signal-to-noise ratio. Then, the MUSIC method takes a long computation time to find many frequencies by increasing the dimension of the autocorrelation matrix. To solve this problem, an algorithm that is based on zooming in a specific frequency range is proposed with MUSIC in order to improve the performances of frequency extraction. Moreover, the method is integrated as a part of MUSIC to estimate the frequency signal dimension order based on classification of autocorrelation matrix eigenvalues. The proposed algorithm has been applied to detect a rotor broken bar fault in a three-phase squirrel-cage induction machine under different loads and in steady-state condition.
314 citations
TL;DR: The findings show that the results have higher precision after the m-D extraction rather than before it, since only the vibrational/rotational components are employed.
Abstract: This paper highlights the extraction of micro-Doppler (m-D) features from radar signal returns of helicopter and human targets using the wavelet transform method incorporated with time-frequency analysis. In order for the extraction of m-D features to be realised, the time domain radar signal is decomposed into a set of components that are represented at different wavelet scales. The components are then reconstructed by applying the inverse wavelet transform. After the separation of m-D features from the target's original radar return, time-frequency analysis is then used to estimate the target's motion parameters. The autocorrelation of the time sequence data is also used to measure motion parameters such as the vibration/rotation rate. The findings show that the results have higher precision after the m-D extraction rather than before it, since only the vibrational/rotational components are employed. This proposed method of m-D extraction has been successfully applied to helicopter and human data.
311 citations
TL;DR: In this paper, the Hilbert-Huang method is presented with modifications, for time-frequency analysis of distorted power quality signals, and the empirical mode decomposition (EMD) is enhanced with masking signals based on fast Fourier transform (FFT), for separating frequencies that lie within an octave.
Abstract: The Hilbert-Huang method is presented with modifications, for time-frequency analysis of distorted power quality signals. The empirical mode decomposition (EMD) is enhanced with masking signals based on fast Fourier transform (FFT), for separating frequencies that lie within an octave. Further, the instantaneous frequency and amplitude of the constituent modes obtained by Hilbert spectral analysis are improved by demodulation. The method shows promising time-frequency-magnitude localization capabilities for distorted power quality signals. The performance of the new technique is compared with that of another multiresolution analysis tool, the S-transform-a phase corrected wavelet transform. Analysis on actual measurements of transformer inrush current from an existing laboratory setup is used to demonstrate this technique.
215 citations
TL;DR: It is shown that increasing the number of crossband filters not necessarily implies a lower steady-state mean-square error (mse) in subbands, and analytical relations between the number and length of the input signal are derived.
Abstract: In this paper, we investigate the influence of crossband filters on a system identifier implemented in the short-time Fourier transform (STFT) domain. We derive analytical relations between the number of crossband filters, which are useful for system identification in the STFT domain, and the power and length of the input signal. We show that increasing the number of crossband filters not necessarily implies a lower steady-state mean-square error (mse) in subbands. The number of useful crossband filters depends on the power ratio between the input signal and the additive noise signal. Furthermore, it depends on the effective length of input signal employed for system identification, which is restricted to enable tracking capability of the algorithm during time variations in the system. As the power of input signal increases or as the time variations in the system become slower, a larger number of crossband filters may be utilized. The proposed subband approach is compared to the conventional fullband approach and to the commonly used subband approach that relies on multiplicative transfer function (MTF) approximation. The comparison is carried out in terms of mse performance and computational complexity. Experimental results verify the theoretical derivations and demonstrate the relations between the number of useful crossband filters and the power and length of the input signal
158 citations
TL;DR: In this paper, a linear discriminant classifier based on the transform coefficients is proposed to detect electrical and mechanical faults in a permanent magnet-magnet ac drive. But the classification of different fault types is not performed.
Abstract: The detection of noncatastrophic faults in conjunction with other factors can be used to determine the remaining life of an electric drive. As the frequency and severity of these faults increase, the working life of the drive decreases, leading to eventual failure. In this paper, methods are presented to identify developing electrical and mechanical faults based on both the short-time Fourier transform and wavelet analysis of the field-oriented currents in permanent-magnet ac drives. The different fault types are classified by developing a linear discriminant classifier based on the transform coefficients.
139 citations
TL;DR: The proposed IF estimator is tested on noisy synthetic monocomponent and multicomponent signals exhibiting linear and nonlinear laws and a classification method using least squares data-fitting is proposed and illustrated on synthetic and real signals.
130 citations
TL;DR: In this paper, the authors proposed the concept of time-frequency entropy based on Hilbert-Huang transform and proposed a gear fault diagnosis method based on timefrequency entropy, which could identify gear status with or without fault accurately and effectively.
123 citations
TL;DR: This paper documents applications of time-variant analysis for damage detection using two main approaches, the time–frequency and the time-scale analyses.
Abstract: Signal processing is one of the most important elements of structural health monitoring. This paper documents applications of time-variant analysis for damage detection. Two main approaches, the time–frequency and the time–scale analyses are discussed. The discussion is illustrated by application examples relevant to damage detection.
121 citations
TL;DR: The analysis and comparisons of the spectrogram, Wigner distribution and wavelet transform techniques to the phonocardiogram signal (PCG) are presented to be able to distinguish the various techniques in their aptitude to separate and present suitably the internal components of these sounds.
95 citations
01 Dec 2007
TL;DR: The amount of samples which must be taken from a signal with sparse Discrete-time Fourier Transform (DFT) can be reduced compared to the original compressed sensing approach if information on the support of the sparse domain can be employed.
Abstract: Compressed sensing has recently emerged as a technique allowing a discrete-time signal with a sparse representation in some domain to be reconstructed with theoretically perfect accuracy from a limited number of linear measurements. Current applications range from sensor networks to tomography and general medical imaging. In this paper, we show that the amount of samples which must be taken from a signal with sparse Discrete-time Fourier Transform (DFT) can be reduced compared to the original compressed sensing approach if information on the support of the sparse domain can be employed. More precisely, the required number of samples in time domain is reduced by exactly the amount of known frequencies associated to non-zero coefficients. Our results additionally provide a link between the so-called fractional Fourier transform and compressed sensing framework, when the positions of all the non-zero components are known.
27 Aug 2007
TL;DR: It is argued that the 2-D Gabor filterbank has the capacity to decompose a patch into its underlying dominant spectro-temporal components, and the response of the filterbank to different speech phenomena is illustrated.
Abstract: We present a 2-D spectro-temporal Gabor filterbank based on the 2-D Fast Fourier Transform, and show how it may be used to analyze localized patches of a spectrogram. We argue that the 2-D Gabor filterbank has the capacity to decompose a patch into its underlying dominant spectro-temporal components, and we illustrate the response of our filterbank to different speech phenomena such as harmonicity, formants, vertical onsets/offsets, noise, and overlapping simultaneous speakers.
TL;DR: In this article, a laboratory grinding spindle-typed rotor bearing system was equipped, and modal testing was carried out to identify system characteristics and operating range, and damage detection from nonstationary mechanical vibration signals collected during acceleration and deceleration was performed.
TL;DR: This paper would like to look for a compromise point between complex mathematics based techniques, such as wavelet packet, sometimes hard to comprehend to the application engineer, and the physical meaning of these tools helping in fixing the real method limits.
TL;DR: A careful study enables us to show that, although both inverses are nearly exact in the infinite continuous domain, this is not true anymore in the practical finite discrete domain.
Abstract: The aim of this paper is to present a study on the potential and limits of the -transform and its inverses. The S-transform is an extension of the short-time Fourier transform with characteristics of the wavelet transform. It is mostly used for time-frequency analyses. Two different inverse S-transforms have been presented in the literature. We explain why the most recent one is an approximation but a very good one. The level of approximation is calculated in this paper. We then discuss the relative merits of both inverses. A careful study enables us to show that, although both inverses are nearly exact in the infinite continuous domain, this is not true anymore in the practical finite discrete domain. Side effects are quantified, and typical examples are given. Time-frequency filtering is one of the main applications of the S-transform. We evaluate the effects that occur when using the S-transform and its inverses for filtering.
TL;DR: A new adaptive multiwindow Gabor expansion is described, which dynamically adapts the windows to the signal's features in time-frequency space, and also yields as by-product an expansion of the signal into layers corresponding to different windows.
Abstract: We describe a new adaptive multiwindow Gabor expansion, which dynamically adapts the windows to the signal's features in time-frequency space. The adaptation is based on local time-frequency sparsity criteria, and also yields as by-product an expansion of the signal into layers corresponding to different windows. As an illustration, we show that simply using two different windows with different sizes leads to decompositions of audio signals into transient and tonal layers. We also discuss potential applications to transient detection and denoising.
TL;DR: In this article, an improved method for computing polarization attributes of particle motion from multicomponent seismic recordings in the time-frequency domain by using the continuous wavelet transform is presented.
Abstract: SUMMARY
We present an improved method for computing polarization attributes of particle motion from multicomponent seismic recordings in the time–frequency domain by using the continuous wavelet transform. This method is based on the analysis of the covariance matrix. We use an approximate analytical formula to compute the elements of the covariance matrix for a time window which is derived from an averaged instantaneous frequency of the multicomponent record. The length of the time-window is automatically and adaptively set to match the dominant period of the analysing wavelet at each time–frequency point. Then the eigenparameters are estimated for each time–frequency point without interpolation. With these key features, our method provides a suitable approach for polarization analysis of dispersive signals or overlapping seismic arrivals in multicomponent seismic data. For polarization analysis in the time domain, we show that the proposed method is consistent with existing polarization analysis methods. We apply the method to real data sets from exploration and earthquake seismology to illustrate some filtering applications and wave type characterizations.
TL;DR: In this paper, the Hilbert-Huang transform (HHT) was adopted to analyze the non-stationary characteristics of wind speed and wind-induced responses of this building under typhoon conditions.
TL;DR: The numerical simulation and experiment have proved the validity of the multiscale windowed Fourier transform for phase extraction of fringe patterns and makes the extracted phase more precise than other methods.
Abstract: A multiscale windowed Fourier transform for phase extraction of fringe patterns is presented. A local stationary length of signal is used to control the window width of a windowed Fourier transform automatically, which is measured by an instantaneous frequency gradient. The instantaneous frequency of the fringe pattern is obtained by detecting the ridge of the wavelet transform. The numerical simulation and experiment have proved the validity of this method. The combination of the windowed Fourier transform and the wavelet transform makes the extracted phase more precise than other methods.
TL;DR: In this article, a new approach for protection of parallel transmission lines is presented using a time-frequency transform known as the S-transform that generates the s-matrix during fault conditions.
Abstract: A new approach for protection of parallel transmission lines is presented using a time-frequency transform known as the S-transform that generates the S-matrix during fault conditions. The S-transform is an extension of the wavelet transform and provides excellent time localisation of voltage and current signals during fault conditions. The change in energy is calculated from the S-matrix of the current signal using signal samples for a period of one cycle. The change in energy in any of the phases of the two lines can be used to identify the faulty phase based on some threshold value. Once the faulty phase is identified the differences in magnitude and phase are utilised to identify the faulty line. For similar types of simultaneous faults on both the lines and external faults beyond the protected zone, where phasor comparison does not work, the impedance to the fault point is calculated from the estimated phasors. The computed phasors are then used to trip the circuit breakers in both lines. The proposed method for transmission-line protection includes all 11 types of shunt faults on one line and also simultaneous faults on both lines. The robustness of the proposed algorithm is tested by adding significant noise to the simulated voltage and current waveforms of a parallel transmission line. A laboratory power network simulator is used for testing the efficacy of the algorithm in a more realistic manner.
TL;DR: In this article, Zhao et al. analyzed time-frequency characteristics of friction-induced vibration and showed that there is always a frequency change in the timefrequency presentation of vibration in the location where the vibration is bounded.
TL;DR: It is demonstrated that the proposed method can successfully separate overlapping targets efficiently and is achieved due to fast Fourier transform-bascd processing, use of a single slice of RAT, and the use of only one-dimensional (1-D) searches.
Abstract: A novel efficient technique based on a single slice Radon-ambiguity transform (RAT) for time-delay and time-scale estimation is proposed. The proposed approach combines the narrowband cross-ambiguity function (NBCAF), the wideband cross-ambiguity function (WBCAF), and a single slice RAT to estimate multiple target parameters in noisy environments. The square modulus of Gaussian-enveloped linear frequency modulated (GLFM) signals has high-energy centrality in the ambiguity plane. Its peaks in the NBCAF fall along nearly straight lines whose slopes depend on the Doppler rates of the moving targets. These lines could be effectively detected by computing the entire Radon transform of the NBCAF for all possible angles; however, it is a computationally intensive procedure. It is shown that without calculating the entire RAT, it is possible to estimate target parameters using only a single slice of the RAT, i.e., using an appropriate projection of the NBCAF. It is demonstrated that the proposed method can successfully separate overlapping targets efficiently. The efficiency is achieved due to fast Fourier transform (FFT)-bascd processing, use of a single slice of RAT, and the use of only one-dimensional (1-D) searches.
TL;DR: In this paper, a smoothed pseudo Wigner-Ville distribution is used to decouple vibration modes completely in order to study each mode separately, which reduces cross-terms which are troublesome in WIGNer-ville distribution and retains the resolution as well.
TL;DR: This paper presents a noise robust feature extraction algorithm NRFE using joint wavelet packet decomposition (WPD) and autoregressive (AR) modeling of a speech signal to improve noise robustness and performance.
01 Jan 2007
TL;DR: In this article, the amplitude of voltage sag was determined by the amplitude characteristics of fundamental frequency voltage and the high frequency signal produced when voltage sag starts and disappears was employed to identify the voltage sag inception and disappearance instant.
Abstract: Short time Fourier transform (STFT) was employed to measure the amplitude of voltage sag, detect interference instant and locate interference source. The amplitude of voltage sag was determined by the amplitude characteristics of fundamental frequency voltage and the high frequency signal produced when voltage sag starts and disappears was employed to identify the voltage sag inception and disappearance instant. A method based on the amplitude of fundamental frequency component and the number of signal singularities is proposed to locate interference source. The method can distinguish the voltage sag induced by fault from that induced by induction motor starting effectively. Simulation tests show that the method has high sensitivity and has better immunity to harmonics and noise than the methods based on wavelet transform.
24 Jun 2007
TL;DR: Based on the regularly spaced pilot pattern, alternative patterns are derived that preserve the non-aliasing property by hopping the scan lines in either time domain or frequency domain, but not both.
Abstract: In an OFDM system, channel estimation can be considered as sampling the time-frequency response of the channel through a number of known pilot symbols placed across the time-frequency plane. Sampling theory dictates that the pilot insertion frequency must be above the Nyquist rates in both time and frequency to avoid aliasing of the delay-Doppler response. Based on the regularly spaced pilot pattern, we can derive alternative patterns that preserve the non-aliasing property by hopping the scan lines in either time domain or frequency domain, but not both. From this extended set of patterns, we find ones with properties that, in addition to channel estimation, can achieve responsively other synchronization tasks such as initial time-frequency offset estimation and device identification. The ambiguity function analysis frequently used in radar signal design leads to a periodic time-hopping pattern based on the costas array that has minimal coincidences with its circular time-frequency shifts, which can be used for the identification of multiple devices. The hopping in time also greatly increases the pilot's time support, thus enabling the quick initial acquisition of timing and frequency offset with very short observation.
TL;DR: Spectral decomposition, by which a time series is transformed from the 1D time/amplitude domain to the 2DTime/spectrum domain, has become a popular and useful tool in seismic exploration for hydrocarbons.
Abstract: Spectral decomposition, by which a time series is transformed from the 1D time/amplitude domain to the 2D time/spectrum domain, has become a popular and useful tool in seismic exploration for hydrocarbons. The windowed, or short-time Fourier transform (STFT) was one early approach to computing the time-frequency (t-f) distribution. This method relies on the user selecting a fixed time window, then computing the Fourier spectrum within the time window while sliding the window along the length of the trace. The primary limitation of the STFT is the fixed window which prevents either time localization of high frequency components (if a long window is used) or spectral resolution of the low-frequency components (if a short window is used).
01 Oct 2007
TL;DR: In this paper, a study of the permanent magnet synchronous motor (PMSM) running under demagnetization has been carried out by means of two dimensional finite element analysis (FEA), and simulations results were compared with experimental results.
Abstract: This paper presents a study of the permanent magnet synchronous motor (PMSM) running under demagnetization. The simulation has been carried out by means of two dimensional (2-D) finite element analysis (FEA), and simulations results were compared with experimental results. The demagnetization fault is analyzed by means of decomposition of stator currents obtained at different speeds. The Hilbert Huang transform (HHT) is used as processing tool. This transformation represents time-dependent series in a two-dimensional (2-D) time-frequency domain by extracting instantaneous frequency components within the signal through an Empirical Mode Decomposition (EMD) process.
TL;DR: In this paper, the authors evaluate the performance of the continuous wavelet transform and empirical mode decomposition in tandem with Hilbert transform in the analysis of a variety of classical nonlinear signals, underscoring a fundamental difference between the two approaches.
Abstract: Recently, there has been growing utilization of time-frequency transformations for the analysis and interpretation of nonlinear and nonstationary signals in a broad spectrum of science and engineering applications. The continuous wavelet transform and empirical mode decomposition in tandem with Hilbert transform have been commonly utilized in such applications, with varying success. This study evaluates the performance of the two approaches in the analysis of a variety of classical nonlinear signals, underscoring a fundamental difference between the two approaches: the instantaneous frequency derived from the Hilbert transform characterizes subcyclic and supercyclic nonlinearities simultaneously, while wavelet-based instantaneous frequency captures supercyclic nonlinearities with a comple- mentary measure of instantaneous bandwidth characterizing subcyclic nonlinearities. This study demonstrates that not only is the spectral content of the wavelet instantaneous bandwidth measure consistent with that of the Hilbert instantaneous frequency, but in the case of the Rossler system, produces identical oscillatory signature.
Patent•
16 Mar 2007TL;DR: In this article, a method and apparatus for performing channel estimation using time-frequency localized pilots and de-noising techniques are disclosed, where a transmitter sends pilot symbols which are localized in a joint timefrequency domain to a receiver for channel estimation, and the receiver performs a timefrequency synthesis, such as an inverse discrete Gabor transform (IDGT), to generate a noise-removed pilot symbols in a time domain.
Abstract: A method and apparatus for performing channel estimation using time-frequency localized pilots and de-noising techniques are disclosed. A transmitter sends pilot symbols which are localized in a joint time-frequency domain to a receiver for channel estimation. The receiver receives the pilot symbols and performs a time-frequency analysis, such as a discrete Gabor transform (DGT), to transform the received pilot symbols to a joint time-frequency domain. The receiver applies a de-noising technique, such as masking, to separate the pilot symbols from the embedded noise in the joint time-frequency domain. The receiver performs a time-frequency synthesis, such as an inverse discrete Gabor transform (IDGT), to generate a noise-removed pilot symbols in a time domain. The noise left after de-noising is only that part that overlaps with the pilot symbols in the joint time-frequency domain. The receiver then performs channel estimation with the noise-removed pilot symbols.