Short Time Fourier Transform Algorithm Research Articles

A study on classifying BSR noise in automotive door trim using a deep learning method

In this paper, we demonstrated a deep learning model that can automatically classifies BSR(Buzz, Squeak, Rattle) noise in automotive door trim using convolutional neural networks. In the preprocessing process, sound features of door trim noises are extracted by using STFT(Short-time Fourier Transform) algorithm to obtain spectrogram images for each noises we obtained from experiment. To classify various noises of door trim, those image data are labeled with its root cause, then it is trained with deep convolutional neural network. Furthermore, t-SNE (t-Stochastic Neighbor Embedding) algorithm was used to visualize the effect of deep convolutional neural network we demonstrated. The classifier of the demonstrated method achieved with the accuracy of 96%.

Asphalt Pavement Transverse Cracking Detection Based on Vehicle Dynamic Response

Transverse cracking is thought of as the typical distress of asphalt pavements. A faster detection technique can provide pavement performance information for maintenance administrations. This paper proposes a novel vehicle-vibration-based method for transverse cracking detection. A theoretical model of a vehicle-cracked pavement vibration system was constructed using the d’Alembert principle. A testing system installed with a vibration sensor was put in and applied to a testing road. Then, parameter optimization of the Short-time Fourier transform (STFT) was conducted. Transverse cracking and normal sections were processed by the optimized STFT algorithm, generating two ideal indicators. The maximum power spectral density and the relative power spectral density, which were extracted from 3D time–frequency maps, performed well. It was found that the power spectral density caused by transverse cracks was above 100 dB/Hz. The power spectral density at normal sections was below 80 dB/Hz. The distribution of the power spectral density for the cracked sections is more discrete than for normal sections. The classification model based on the above two indicators had an accuracy, true positive rate, and false positive rate of 94.96%, 92.86%, and 4.80%, respectively. The proposed vehicle-vibration-based method is capable of accurately detecting pavement transverse cracking.

Differentiable Short-Time Fourier Transform Window Length Selection Driven by Cyclo-Stationarity

The Short-Time Fourier transform is widely applied in the condition monitoring of rotating machinery. Even so, selecting the optimal window length for the Short-Time Fourier Transform remains a challenge. This work presents a procedure for adapting the Short Time Fourier Transform algorithm to be differentiable with respect to window length by using continuous window functions defined over the entire input signal duration. Thanks to this modification, a differentiable loss criterion can be defined to measure the Short-Time Fourier quality, and the gradient of the loss criterion with respect to window length can be computed. The optimal window length for a given loss criterion can then be efficiently solved for using a gradient-based optimization algorithm. Results from a simulated bearing dataset and three experimental bearing datasets are used to compare the optimal spectrograms obtained using different loss criteria. Specifically, a sparsity-based loss criterion is compared with two loss criteria inspired by the characteristic cyclo-stationarity machine of faults in rotating machinery. The results demonstrate the effectiveness of the differentiable window length selection method and highlight the importance of selecting appropriate loss criteria for defining STFT quality. Loss criteria that account for the cyclo-stationary nature of the signals are shown to be less likely to target single high-amplitude impulsive events compared to the sparsity-based loss criterion.

Relationship between Twin-to-Twin Selective Intrauterine Growth Restriction with sFas/sFasL Level of Umbilical Cord Blood Using Doppler Ultrasound Fetal Heart Rhythm Detection Algorithm

The study focused on the application value of ultrasound Doppler fetal heart rate detection algorithm based on short-time Fourier transform (STFT) in the diagnosis of twin-to-twin selective intrauterine growth restriction (sIUGR) and the correlation between twin-to-twin sIUGR and sFas/sFasL levels of umbilical cord blood. The normalized method was introduced into the STFT algorithm to optimize it to detect the fetal instantaneous heart rate. 82 pregnant women with twin pregnancies were selected as the research subjects and they were divided into the restricted group (41 cases) and the control group (41 cases) according to whether the fetus had selective growth restriction. The two groups were compared for the differences in the fetal mortality, complication rate, and sFas/sFasL expression levels. The results showed that the STFT-based ultrasonic Doppler fetal heart rate detection algorithm could ensure the quality of the fetal heart rate signal and had high resolution at the 200–400 Hz characteristic frequency band and that the accuracy in distinguishing S1 and S2 was 5.8% higher than that of the traditional autocorrelation algorithm. The proportion of abnormal fetal heart rate in the restricted group was significantly higher than that in the control group ( P < 0.05 ), birth weight was significantly lower than that in the control group ( P < 0.05 ), and fetal mortality was significantly higher than that in the control group ( P < 0.05 ). There was no statistical difference in the incidence of complications between the two groups ( P > 0.05 ). In restricted group, the content of sFas in cord blood was (3326.54 ± 317.42) pg/mL and that in the control group was (2003.29 ± 196.45) pg/mL. The content of sFas in cord blood of the restricted group was significantly higher than that of the control group ( P < 0.01 ). In the restricted group, the content of sFasL in cord blood was (382.52 ± 36.17) pg/mL, and that in the control group was (180.84 ± 16.20) pg/mL. The content of sFasL in cord blood of the restricted group was significantly higher than that of the control group ( P < 0.001 ). It was concluded that the STFT-based ultrasound Doppler fetal heart rate monitoring is beneficial to early diagnosis and timely intervention of twin-to-twin sIUGR.

Short-Time Fourier Transform Based on Metaprogramming and the Stockham Optimization Method.

The extension for high-performance STFT (Short-Time Fourier Transform) algorithm written entirely in Java language for non-parallel computations is presented in the current paper. This solution could compete with the best available and most common algorithms supplied by libraries such as FFTW or JTransform. The main idea was to move complex computations and expensive functions to the program generation phase. Thus, only core and essential operations were executed during the runtime phase. Furthermore, new approach allows to eliminate the necessity for a rearrangement operation that uses the bit-reversal permutation technique. This article presents a brief description of the STFT solution that was worked out as an extension for the original application, in order to increase its efficiency. The solution remains a Stockham algorithm adapted using metaprogramming techniques and entails an additional reduction its execution time. Performance tests and experiments were conducted using a Java Platform and JMH library, which allowed for accurate execution time measurements. Major aspects of the Java VM like warm-up effects were also taken into consideration. Solution was applied into Electrical Capacitance Tomography measurement system in order to measure the material changes during the silo discharging industrial process.

RETRACTED: FPGA medical big data system and ischemic stroke rehabilitation nursing

Ischemic stroke is one of the most deadly illnesses in the world, leading to high mortality. Due to lung disease, stroke is the abnormal growth of cells characterized by a single irregular cell and spreads throughout the body. Therefore, to detect and heal the affected area at an early stage, it is necessary to detect the affected area after application. Ischemic stroke is generally regarded as an essential indicator of stroke rehabilitation care. The previous method uses SVM (Support Vector Machine) and STFT (Short Time Fourier Transform Algorithm) to process an image processing system based on stroke detection. This is more accurate and efficient for CT (Computed Tomography) images. The conversion method is significantly slower, and the advanced risk architecture cannot verify the image. The proposed FPGA (Field Programmable Gate Array) and CNN (Convolutional Neural Network) are used to develop image processing and easily interact with the database without introducing complexity. FPGA (Field Programmable Gate Array) is mainly realized by ASIC (Application Specific Integrated Circuit). The system speeds up detecting strokes and lung diseases and can be used as a single process system or another biomedical imaging system component. According to the medical big data system, the image processing system relies on bilateral filtering, edge detection, multiple thresholds, image segmentation, morphological image processing, and image labeling to collect stroke symptoms.

STFT spectrogram based hybrid evaluation method for rotating machine transient vibration analysis

The main purpose of this paper is to represent a methodenabling vibration components to be extracted from a high-resolution Short-Time Fourier-Transformation (STFT) based spectrogram assessed as an image to support transient analysis on rotating machines. Therefore, an improved STFT algorithm was developed to allocate and utilize computational memory more efficiently. The resulting spectrogram was compressed into a grey-scale image without any kind of information loss and was used for further image processing methods to obtain details about the vibration components. Furthermore, differential and moving average predictive tracking algorithms were developed for frequency ridge evaluation in the spectrogram image. For further analysis, the obtained results were transformed back with an inverse transformation method from image-space to time–frequency plane. Moreover, these results are able to be used to estimate the speed of rotation of the machine and to observe the frequency components. The methods were tested and validated with simulated signals and transient measurements on rotating machines. With the combination of vibration signal- and image processing techniques the evaluation time and computational resource requirements are decreased enhancing more efficient and accurate analysis, nevertheless opens the possibility of a real-time condition monitoring based on a basic vibration measurement.

RETRACTED: Urban ecological forestry analysis based on FPGA and dynamic image sampling

Abstract Urban ecological forestry is one of the analyses leading to high global mortality. Urban ecological forestry is based on the destruction of forest habitats and the loss of biodiversity. It is the growth of animals and plants in the world. It is characterized by a single irregular urban forest all over the world. Therefore, it is necessary to detect the affected area according to the application, find out and deal with it early. It is generally believed that urban ecosystem forests are essential indicators of destruction. The previous method used Support Vector Machine (SVM) and Short Time Fourier Transform Algorithm (STFT) to reduce the accuracy and efficiency of C.T. (Computed Tomography) images. The image processing system based on urban ecological forest analysis and detection is being processed. The conversion method is prolonged, and it is impossible to verify the image using advanced risk architecture. The proposedProgrammable Gate Array) is used to develop image processing and easily interface data without complexity. Programmable Gate Array is mainly realized by ASIC (Application Specific Integrated Circuit). This system accelerates forest detection and loss detection and can be used as a single processing system or as a component of an image detection system. Image processing systems rely on double-sided filtering, edge detection, multiple thresholds, image segmentation, morphological image processing, and image labeling to solve urban ecological forestry analysis symptoms.

Dedicated processor for hologram calculation using sparse Fourier bases.

Recently, a calculation method involving sparse point spread functions in the short-time Fourier transform (STFT) domain was proposed. In this paper, a dedicated processor using the STFT algorithm is described, which is implemented on a field-programmable gate array. All the operations in this algorithm are implemented using fixed-point arithmetic. Since this algorithm includes a trigonometric function and an error function, lookup tables (LUTs) are utilized to reduce the calculation costs. We have devised a dedicated circuit architecture that allows parallel operations. In addition, a central processing unit could generate holograms using the STFT-based algorithm with fixed-point arithmetic and LUTs at a higher speed than the generation using floating-point arithmetic.

Endpoint detection of distributed fiber sensing systems based on STFT algorithm

This paper proposes an endpoint detection method based on short-time Fourier transform (STFT), which can be applied to fiber optic interferometric vibration sensing systems. This scheme consists of time-frequency analysis, feature extraction and thresholds judgment. Field experimental results demonstrate that the average error of 100 experiments of the proposed scheme is lower than the error of the short-time zero-crossing ratio method. The mean processing time (0.149 s) is lower than the sampling time (0.3 s), which allows the system to perform continuous in-field intrusion detection related applications.

SPECTRAL DECOMPOSITION TECHNIQUE BASED ON STFT AND CWT FOR IDENTIFYING THE HYDROCARBON RESERVOIR

The spectral decomposition is one of the advanced interpretation techniques such as seismic inversion, amplitude versus offset analysis, and seismic attribute that helpful in direct interpretative approach in seismic exploration. This technique is a transformation algorithm, thus a signal can be transformed into its varying frequency contained in the seismic signal. There are a variety of spectral decomposition algorithms in the decomposing seismic signal from time domain into frequency domain. These algorithms include Short Time Fourier Transform (STFT) and Continuous Wavelet Transform (CWT). The STFT algorithm is a conventional and simple technique for computing a time-frequency spectrum, which is based on the application of Fourier transform. However, the STFT algorithm has a problem related to the frequency resolution. In its implementation, this algorithm is limited by predefi ned window length. In contrast, the CWT algorithm is believed to be able to overcome the limitation of window length. The CWT threats wavelet at certain window length, which is defi ned by the characteristics of the wavelet. In this study, the comparison between spectral decomposition technique based on STFT and CWT method was performed, particularly in its application to the synthetic and real data set. Each algorithm has its own advantages and disadvantages in decomposing the seismic signal. Further, this analysis can be used as a reference to select one of two algorithms for the specifi c application. The synthetic data set application shows that CWT algorithm produces better frequency resolution compared to STFT algorithm. In addition, the real data set application shows that time frequency section of the seismic line provides a spectral feature, which is useful to identify the hydrocarbon reservoir, which is associated with low-frequency shadow zone.

Short-Time Fourier Transform Based Transient Analysis of VSC Interfaced Point-to-Point DC System

The transient response of the voltage source converter (VSC) interfaced dc system is significantly different from the ac counterpart. The rapid discharge current from the dc-link capacitors and the vulnerability of the freewheeling diodes during the short circuit in dc grid demand that the transient detection algorithm executes within a few milliseconds. The rapidly rising fault current in the dc grid is expected to have high-frequency components which might be an effective indicator of the transient condition. This paper presents quantitative investigation of the high-frequency components utilizing short-time Fourier transform (STFT) during transient conditions. Detailed operating principles with various factors affecting the STFT operation such as ripple content of the input dc signal and window type and length have been thoroughly investigated. STFT algorithm is able to detect low-impedance faults within 1 ms and high-impedance faults in 2 ms. Moreover, it is able to distinguish between short-circuit fault and less severe transient conditions such as sudden load change. The STFT algorithm is evaluated analytically and subsequently applied to a MATLAB/Simulink based dc test system. It is further validated and substantiated with the real fault current data obtained from a scaled-down experimental testbed. Sensitivity analysis and comparison with the existing frequency-domain-based fault-detection method are done to support the efficacy of the proposed method.

Power System Dynamic Frequency Measurement Based on Novel Interpolated STFT Algorithm

As the product of time and frequency resolution of the Short-Time Fourier Transform (STFT) constant, the dynamic frequency measurement accuracy provided by the STFT is reduced under asynchronous sampling. This paper proposes a novel Interpolated Short-Time Fourier Transform (IpSTFT) based on the Blackman window. The Blackman window is adopted to reduce the spectral leakage, and the spectral interpolation procedure is applied to eliminate the picket-fence effects. Results of simulations are provided to show that the proposed method can improve the accuracy of dynamic frequency measurement significantly with low time consumptions.

P 105 Spatio-temporal characteristics of somatosensory-evoked high-gamma activity of the EEG

Introduction There is enlarging interest in the cortical function of high gamma band (HGB) changes related to different sensory modalities, movement tasks, and higher cognitive processes. EEG experiments have shown visually-evoked, auditory-evoked, and movement-related HGB changes, including the range of 50–70 Hz ( Tallon-Baudry et al., 1997 , Ball et al., 2008 ). While HGB activity in this frequency range associated to peripheral somatosensory stimulation was reported for intracranial recordings in neurological patients ( Fukuda et al., 2008 ), it has received little attention in EEG studies. Here the focus was on very high frequency somatosensory responses above 600 Hz ( Curio et al., 1994 ). In the present study, we implemented an experimental setup, optimized to measure high frequencies in the EEG, to investigate if robust 50–70 Hz HGB activity in noninvasive EEG could be identified during median nerve SEP stimulation and what the spatio-temporal characteristics of these changes are. Methods In an electromagnetically-shielded cabin, EEG data of 15 healthy subjects was obtained with 128 Ag/AgCl scalp electrodes and with NeurOne amplifiers (24-bit digital resolution and 5 kHz sampling rate). 1200 SEPs per subject were evoked by stimulation of the left median nerve with a SEP-stimulator. Simultaneous eye-tracking data from 7 of these subjects was attained to examine the influence of ocular movements on the neuronal recordings. Data analysis included preprocessing, short-time Fourier transformation algorithm, normalization, calculation of induced and evoked responses and a false discovery rate corrected sign test. Results All analyzed subjects showed a significant spectral power increase in the HGB range at 50–70 Hz after median nerve stimulation ( p Conclusion Using our purpose-build experimental set up, we were able to demonstrate a specific contralateral somatosensory-evoked response pattern in the 50–70-Hz frequency range of the noninvasive EEG. Our results agree with the findings of previous intracranial SEP studies and open up new possibilities to noninvasively study sensory related brain functions, both in healthy subjects and neurological patients.

Semi-active control of vibrations of spar type floating offshore wind turbines

A semi-active algorithm for edgewise vibration control of the spar-type floating offshore wind turbine (SFOWT) blades, nacelle and spar platform is developed in this paper. A tuned mass damper (TMD) is placed in each blade, in the nacelle and on the spar to control the vibrations for these components. A Short Time Fourier Transform algorithm is used for semi-active control of the TMDs. The mathematical formulation of the integrated SFOWT-TMDs system is derived by using Euler-Lagrangian equations. The theoretical model derived is a time-varying system considering the aerodynamic properties of the blade, variable mass and stiffness per unit length, gravity, the interactions among the blades, nacelle, spar, mooring system and the TMDs, the hydrodynamic effects, the restoring moment and the buoyancy force. The aerodynamic loads on the nacelle and the spar due to their coupling with the blades are also considered. The effectiveness of the semi-active TMDs is investigated in the numerical examples where the mooring cable tension, rotor speed and the blade stiffness are varying over time. Except for excessively large strokes of the nacelle TMD, the semi-active algorithm is considerably more effective than the passive one in all cases and its effectiveness is restricted by the low-frequency nature of the nacelle and the spar responses.

Comment on "Quantitative comparison of analysis methods for spectroscopic optical coherence tomography".

In a recent paper by Bosschaart et al. [Biomed. Opt. Express 4, 2570 (2013)] various algorithms of time-frequency signal analysis have been tested for their performance in blood analysis with spectroscopic optical coherence tomography (sOCT). The measurement of hemoglobin concentration and oxygen saturation based on blood absorption spectra have been considered. Short time Fourier transform (STFT) was found as the best method for the measurement of blood absorption spectra. STFT was superior to other methods, such as dual window Fourier transform. However, the algorithm proposed by Bosschaart et al. significantly underestimates values of blood oxygen saturation. In this comment we show that this problem can be solved by thorough design of STFT algorithm. It requires the usage of a non-gaussian shape of STFT window that may lead to an excellent reconstruction of blood absorption spectra from OCT interferograms. Our study shows that sOCT can be potentially used for estimating oxygen saturation of blood with the accuracy below 1% and the spatial resolution of OCT image better than 20 μm.

Feasibility of MFC (Macro-Fiber Composite) Transducers for Guided Wave Technique

Since MFC(macro-fiber composite) transducer has been developed, many researchers have tried to apply this transducer on SHM(structural health monitoring), because it is so flexible and durable that it can be easily embedded to various kinds of structures. The objective of this paper is to figure out the benefits and feasibility of applying MFC transducers to guided wave technique. For this, we have experimentally tested the performance of MFC patches as transmitter and sensors for excitation and reception of guided waves on the thin aluminum alloy plate. In order to enhance the signal accuracy, we applied the FIR filter for noise reduction as well as used STFT(short-time Fourier transform) algorithm to image the guided wave characteristics clearly. From the results, the guided wave generated based on MFC showed good agreement with its theoretical dispersion curves. Moreover, the ultrasonic Lamb wave techniques based on MFC patches in pitch-catch manner was tested for detection of surface notch defects of which depths are 10%, 20%, 30% and 40% of the aluminum plate thickness. Results showed that the notch was detectable well when the notch depth was 10% of the thickness or greater.

The FWB Filter with Real-Time Detector to Suppress Category II Pilot Induced Oscillation

PIO (Pilot Induced Oscillation) is the key factor to damage the flight safety and decrease the quality of manned aircraft. Most of modern PIO records associating with rate limiting or position limiting from actuation systems are defined Category II PIO. The existing compensators, e.g. the Feedback-with-Bypass (FWB) filter, can reduced the oscillation potentiality of the pilot-in-the-loop aircraft system due to rate limiting of control surface, however a negative effect evidently arose which slow the response speed of whole system. This paper uses a Real-Time PIO detector based on Short-time Fourier Transform algorithm as a trigger to determine the switch-off time of the compensator working on the whole system so as to suppress the negative effect on responses of system by FWB. Simulation results show us the positive conclusion that this kind combined FWB filter with Real-Time PIO detector is effectively preventing the Category II PIO without performance decreasing.

Base Isolation Systems Employing Variable Friction Dampers Based on a STFT Controller

Conventional isolation systems may induce an excessive response in near-fault earthquakes. A new short time Fourier transformation (STFT) control algorithm for variable friction dampers (VFD) is developed to improve the performance of base isolation buildings in near-fault earthquakes. The STFT controller varies the clamping force in the VFD damper to achieve the response reduction. In addition, the STFT algorithm is implemented analytically on a multi degree of freedom system (MDOF) with laminated rubber bearings and variable friction dampers in Simulink environment. Three types of earthquakes representing a wide variety of ground motions are considered as the ground excitations in the simulation. The numerical show that, compared with conventional isolation systems, the semi-active isolation system controlled by the STFT algorithm can reduce the excessive response in near-fault earthquakes effectively.

Acoustic emission in bearing steel during isothermal formation of midrib

The first attempt of an acoustic emission (AE) application for the control of isothermal transformation, at the temperature near the martensite start (M S ) in 100CrMnSi6-4 bearing steel, is made herein. A unique method for recording the signals of the AE was developed. The Short Time Fourier Transform algorithm was used for displaying the signal values on the spectrogram graphs. Acoustic emission showed the nucleation and growth of the isothermal martensite in the residual austenite. The maximum power spectral density of signals was determined using the relative intensity of the RMS values. The observed energy of separated signals during the second stage of residual austenite decomposition is comparable to the energy of the initial stage at lower temperatures. It was proposed as a new process concept under paraequilibrium that involves the midrib formation and swing back in the kinetics of lower bainite.