Time–frequency analysis

About: Time–frequency analysis is a research topic. Over the lifetime, 5407 publications have been published within this topic receiving 104346 citations.

Development of a time–frequency approach to quantify railroad ballast fouling condition using ultra-wide band ground-penetrating radar data

Abstract: This paper discusses the use of ground-penetrating radar (GPR) to assess railroad track substructure conditions. An ultra-wide band (UWB) GPR system, having a centre frequency at or higher than 2 GHz, can be used to detect the scattering pattern and to predict air void volume in railroad ballast. A time–frequency technique was implemented to characterise the signal in time and frequency domains simultaneously. Because electromagnetic energy attenuation is highly frequency dependent, the frequency sub-bands of the reflected UWB GPR signal can be analysed separately to quantify the fouling material and quantify moisture content. Additionally, to validate the GPR system capability, a ground truth field survey was conducted. Using ballast samples collected from the field for validation, this paper shows that a time–frequency analysis may provide a new method to measure the thickness of clean ballast, detect the trapped water and assess the ballast fouling and moisture content along the track.

Bearing fault detection and diagnosis based on teager–huang transform

Abstract: A new approach to fault diagnosis of bearings based on the Teager–Huang Transform (THT) is presented. This method is based on the Empirical Mode Decomposition (EMD) and Teager Kaiser Energy Operator (TKEO) techniques. EMD can adaptively decompose the vibration signal into a series of zero mean Amplitude Modulation-Frequency Modulation (AM-FM) Intrinsic Mode Functions (IMFs). TKEO can track the instantaneous amplitude and instantaneous frequency of the AM-FM component at any instant. The experimental examples are conducted to evaluate the effectiveness of the proposed approach. The experimental results provide strong evidence that the performance of the Teager–Huang Transform approach is better than that of the Hilbert–Huang Transform approach for bearing fault detection. The Teager–Huang Transform has better resolution than the Hilbert–Huang Transform. The Teager–Huang Transform can effectively diagnose the faults of the bearing, thus providing a viable processing tool for gearbox defect monitoring.

Adaptive System Identification in the Short-Time Fourier Transform Domain Using Cross-Multiplicative Transfer Function Approximation

Abstract: In this paper, we introduce cross-multiplicative transfer function (CMTF) approximation for modeling linear systems in the short-time Fourier transform (STFT) domain. We assume that the transfer function can be represented by cross-multiplicative terms between distinct subbands. We investigate the influence of cross-terms on a system identifier implemented in the STFT domain and derive analytical relations between the noise level, data length, and number of cross-multiplicative terms, which are useful for system identification. As more data becomes available or as the noise level decreases, additional cross-terms should be considered and estimated to attain the minimal mean-square error (mse). A substantial improvement in performance is then achieved over the conventional multiplicative transfer function (MTF) approximation. Furthermore, we derive explicit expressions for the transient and steady-state mse performances obtained by adaptively estimating the cross-terms. As more cross-terms are estimated, a lower steady-state mse is achieved, but the algorithm then suffers from slower convergence. Experimental results validate the theoretical derivations and demonstrate the effectiveness of the proposed approach as applied to acoustic echo cancellation.

Identification and Extraction of Surface Waves from Three‐Component Seismograms Based on the Normalized Inner Product

Abstract: Identification of different wave types in a seismogram is an important step for the understanding of wave propagation phenomena. Because in most seismograms, different types of waves with different frequencies may appear simultaneously, separation of waves is more effectively achieved when a time–frequency analysis is performed. In this work, we propose a new time–frequency analysis procedure to identify and extract Rayleigh and Love waves from three‐component seismograms. Exploiting the advantage of the absolute phase preservation by the Stockwell transform, we construct time–frequency filters to extract waves based on the normalized inner product (NIP). Because the NIP is the time–frequency counterpart of the correlation, Rayleigh and Love waves can be identified depending on the NIP between the Stockwell transforms of the horizontal and vertical displacement components. The novelty and advantage of the proposed procedure is that it does not require specifying a priori the direction of propagation of the surface waves, but instead such direction is determined. Furthermore, it is shown that the NIP is a more stable parameter in the time–frequency domain when compared to the instantaneous reciprocal ellipticity, and thus it avoids smoothing (and with it, altering) the data. The procedure has been successfully tested with real signals, specifically to extract Rayleigh and Love waves from seismograms of one aftershock of the 1999 Chi‐Chi earthquake. With the proposed procedure, we found different directions of propagation for retrograde and prograde Rayleigh waves, which might suggest that they are generated by different mechanisms.