Signals In Time-frequency Domain Research Articles

Rapid identification of time-frequency domain gravitational wave signals from binary black holes using deep learning

Abstract Recent developments in deep learning techniques have offered an alternative and complementary approach to traditional matched filtering methods for the identification of gravitational wave (GW) signals. The rapid and accurate identification of GW signals is crucial for the progress of GW physics and multi-messenger astronomy, particularly in light of the upcoming fourth and fifth observing runs of LIGO-Virgo-KAGRA. In this work, we use the 2D U-Net algorithm to identify the time-frequency domain GW signals from stellar-mass binary black hole (BBH) mergers. We simulate BBH mergers with component masses from 7 to 50 $M_{\odot}$ and account for the LIGO detector noise. We find that the GW events in the first and second observation runs could all be clearly and rapidly identified. For the third observing run, about 80\% GW events could be identified. Compared to the traditional convolutional neural network, the U-Net algorithm can output time-frequency domain signal images corresponding to probabilities, providing a more intuitive analysis. In conclusion, the U-Net algorithm can rapidly identify the time-frequency domain GW signals from BBH mergers.

Evaluation of grouting effectiveness for Semi-Rigid pavement base layer cracks based on Time-Frequency domain signal characteristics of 3D GPR

This study aims to address the deficiency of rapid, comprehensive, and non-destructive methods for evaluating the effectiveness of grouting treatment on semi-rigid pavement base layer cracks. By integrating radar simulation with empirical measurements, we establish a comprehensive evaluation method for the grouting treatment effectiveness of pavement base layer cracks. Initially, we propose a GprMax simulation of crack models with varying grout-filled volumes, analyzing the time–frequency domain signal characteristics before and after grouting. Based on data sensitivity, thirteen time–frequency domain signal characteristics are selected as grading evaluation indicators. Subsequently, relying on an engineering case study, we conduct a graded evaluation of the grouting effect on base layer cracks, analyzing the correlation and weight of evaluation indicators to derive a comprehensive evaluation method. Field coring results confirm the accuracy of the comprehensive graded evaluation method.

Research on tool wear prediction based on the random forest optimized by NGO algorithm

Tool wear, which affects both machining quality and manufacturing efficiency, is a major factor in restricting the development of machining intelligence during the machining process. The present study introduces a novel approach that makes use of the northern goshawk optimization (NGO) algorithm for optimizing the random forest (RF) to predict tool wear. First, an analysis of the vibration signal during the cutting process is conducted in both the time-domain and frequency-domain. Next, the S-transform is introduced to analyze the signal in time-frequency domain. Then, the minimal redundancy maximal relevance (mRMR) algorithm is used to screen the sensitive features of tool wear. Finally, milling experiments demonstrate the feasibility of the proposed method and can improve prediction accuracy. Compared with the evaluation indicators of RF, RF optimized by Genetic Algorithm (GA) and RF model prediction results by Gray Wolf Optimization (GWO), this method has higher prediction accuracy and better model fitting accuracy. The results of the study provide technical and theoretical support for the implementation of online intelligent tool wear monitoring, meanwhile for the advancement of cutting automation and the implementation of intelligent tool control.

DTP-Net: Learning to Reconstruct EEG Signals in Time-Frequency Domain by Multi-Scale Feature Reuse.

Electroencephalography (EEG) signals are prone to contamination by noise, such as ocular and muscle artifacts. Minimizing these artifacts is crucial for EEG-based downstream applications like disease diagnosis and brain-computer interface (BCI). This paper presents a new EEG denoising model, DTP-Net. It is a fully convolutional neural network comprising Densely-connected Temporal Pyramids (DTPs) placed between two learnable time-frequency transformations. In the time-frequency domain, DTPs facilitate efficient propagation of multi-scale features extracted from EEG signals of any length, leading to effective noise reduction. Comprehensive experiments on two public semi-simulated datasets demonstrate that the proposed DTP-Net consistently outperforms existing state-of-the-art methods on metrics including relative root mean square error (RRMSE) and signal-to-noise ratio improvement ( ∆SNR). Moreover, the proposed DTP-Net is applied to a BCI classification task, yielding an improvement of up to 5.55% in accuracy. This confirms the potential of DTP-Net for applications in the fields of EEG-based neuroscience and neuro-engineering. An in-depth analysis further illustrates the representation learning behavior of each module in DTP-Net, demonstrating its robustness and reliability.

A proposed tacho-less order tracking technique suitable for railway vehicle condition monitoring applications

To perform a condition monitoring for rotating machines with variable running speed, order tracking is considered as an effective technique suitable for their non-stationary vibration signals. A tachometer signal is necessary to perform a computed order tracking (COT) technique; however, in some applications it's not possible to add tachometer hardware due to its complexity or shortage of power. To improve the applicability of order tracking techniques, some tacho-less order tracking techniques have been introduced in the last few years. These techniques are working off-line by inspecting the vibration signals in time-frequency domain to extract the machine rotating speed. In this paper, a new tacho-less technique suitable for continuous structural health monitoring applications of railway vehicles, is proposed. In the presented approach, a coarse angular speed of the railway axle is estimated firstly using estimation train speed. Then this coarse angular speed is used to improve the applicability of the traditional tacho-less techniques to be able to work online. The train speed is estimated from two acceleration signals obtained from vertical vibration of axle-boxes from same side of train. The sequential properties of the two signals are used to estimate the delay time of one signal with respect to the other one by means of cross-correlation function. By knowing the horizontal distance between the two sensors collected the vibration signals, the train speed can be estimated. This modification of tacho-less technique improves the effectiveness and robustness of the traditional maxima tracking strategy to be performed automatically. Thereafter, signal processing technique based on the computed order tracking as well as averaging technique can be applied to extract the useful information from noisy and non-stationary signals. This technique has been proposed in a framework of a project investigating crack detection using vibration measurements, therefore, vibration signals of cracked axle have been processed using tacho-based method and using the proposed tacho-less technique and a good agreement has been found between the two techniques.

A Novel Time-frequency Domain Reflectometry Method Based on S-transform for Power Cable Defect Localization

Background:: The large-scale aging of cross-linked polyethylene (XPLE) leads to an increase in the frequency of power cable accidents, which poses a threat to the safe operation of the power grid. Objective:: Traditional time-frequency domain reflectometry (TFDR) techniques for fault localization in power cables are susceptible to cross-terms interference, which makes it impossible to accurately locate defects. To overcome the limitation, this paper has proposed a novel cable fault localization method based on S-transform. Method:: The method employs S-transform to perform time-frequency analysis on the collected signals of TFDR, mapping the time-domain signals into time-frequency domain signals. Then, cross-correlation of the time-frequency domain signals is estimated, allowing for accurate positioning of cable defects. Results:: The results of the field experiments conducted on a 10 kV cross-linked polyethylene cable with artificially introduced defects have demonstrated that the proposed method could effectively locate the cable defects with small absolute error. Conclusion:: Compared to existing TFDR methods, the proposed method has been found to be free from interference of cross-terms, resulting in higher reliability and smaller blind spots for detecting cable defects.

The use and effective analysis of vocal spectrum analysis method in vocal music teaching

Abstract As computer science and technology continue to evolve and become more pervasive, their application in analyzing the audio spectrum of vocalizations offers valuable insights for vocal music education. This study introduces a method utilizing Fourier transform analysis to examine time-frequency domain signals in vocal teaching. Initially, voice frequencies are collected during vocal music instruction. Subsequently, these frequencies are processed to extract characteristic sequences, which are then reduced in scale to develop a model for voice spectrum recognition tailored to vocal music education. This model facilitates detailed spectral analysis, enabling the investigation of its auxiliary benefits in vocal music teaching, particularly in identifying prevalent instructional challenges. Our findings indicate that during training on vowels “a” and “i,” professional singers’ pitch at 4kHz declined to between −15 and −18 dB, whereas students’ pitch varied around ±6dB, trending upwards. In cases of air leakage, significant gaps were observed at frequencies of 5500Hz, 10500Hz, and 14500Hz. At the same time, students exhibited missing frequencies at 7kHz, 12kHz, and 14kHz during glottal tone production, with pronounced, abrupt peaks occurring when vocal folds were tightly constricted and devoid of excessive links. This research substantiates the theoretical and practical benefits of digital spectrum technology in enhancing vocal music education, thereby providing a scientific and supportive role.

Delay-Doppler-Based Key Generation Using OTFS

Key generation for secure communication is challenging in high-mobility scenarios due to the low coherence time. This is further exacerbated in frequency-division duplex systems due to the non-reciprocal wireless channel in uplink and downlink. The current work leverages the symplectic Fourier transform to convert fast-varying time-frequency domain signals to slow-varying delay-Doppler (DD) domain and utilizes the wireless channel representation in the DD domain to generate shared keys. The proposed approach is evaluated in terms of key generation and mismatch rates considering correlated eavesdropper, as well as the various randomness criteria provided by the National Institute of Standards and Technology.

Supervised Single Channel Speech Enhancement Method Using UNET

This paper proposes an innovative single-channel supervised speech enhancement (SE) method based on UNET, a convolutional neural network (CNN) architecture that expands on a few changes in the basic CNN architecture. In the training phase, short-time Fourier transform (STFT) is exploited on the noisy time domain signal to build a noisy time-frequency domain signal which is called a complex noisy matrix. We take the real and imaginary parts of the complex noisy matrix and concatenate both of them to form the noisy concatenated matrix. We apply UNET to the noisy concatenated matrix for extracting speech components and train the CNN model. In the testing phase, the same procedure is applied to the noisy time-domain signal as in the training phase in order to construct another noisy concatenated matrix that can be tested using a pre-trained or saved model in order to construct an enhanced concatenated matrix. Finally, from the enhanced concatenated matrix, we separate both the imaginary and real parts to form an enhanced complex matrix. Magnitude and phase are then extracted from the newly created enhanced complex matrix. By using that magnitude and phase, the inverse STFT (ISTFT) can generate the enhanced speech signal. Utilizing the IEEE databases and various types of noise, including stationary and non-stationary noise, the proposed method is evaluated. Comparing the exploratory results of the proposed algorithm to the other five methods of STFT, sparse non-negative matrix factorization (SNMF), dual-tree complex wavelet transform (DTCWT)-SNMF, DTCWT-STFT-SNMF, STFT-convolutional denoising auto encoder (CDAE) and casual multi-head attention mechanism (CMAM) for speech enhancement, we determine that the proposed algorithm generally improves speech quality and intelligibility at all considered signal-to-noise ratios (SNRs). The suggested approach performs better than the other five competing algorithms in every evaluation metric.

A novel bearing fault diagnosis method under small samples using time-frequency multi-scale convolution layer and hybrid attention mechanism module

Deep neural networks for bearing fault diagnosis have become the focus of research in recent years with its excellent feature extraction capability. However, the problem of diagnosis under small samples still needs to be solved in industrial applications, because bearings rarely work in the fault state in practice, resulting in the scarcity of fault data. To solve this problem, this paper proposes a new diagnosis model, a time-frequency multi-scale attention network, which structure allows the original signal and its transformed spectrum to be used as the input in parallel. A multi-scale convolutional layer is also designed to extract information from the signal at different scales to enhance the feature extraction capability of the network. In addition, a hybrid attention mechanism is added to integrate the redundant features and realize the complementarity between features. The experimental results of seven bearing diagnosis cases from two bearings show that the proposed method can achieve high diagnostic accuracy under small samples, which proves the superiority of the proposed method. The time domain signal and frequency domain signal were respectively used as input to train the model. By comparing the accuracy with the time-frequency combined signal as input, the superiority of the time-frequency domain signal as input is proved.

Ubiquitous Domain Adaptation at the Edge for Vibration-Based Machine Status Monitoring

This article presents a ubiquitous domain adaptation (UDA) and generalizability technique for vibration-based automated machine status monitoring at the edge. The method significantly reduces the effects of signal noise artifacts and device/usage-specific vibration signatures using basic time-frequency domain signal operations and a lightweight ensemble of data-driven classifiers, allowing the method to be used for reliable domain-invariant status monitoring of motorized equipment. An experimental setup using vibration data from an air-cooled electric blender motor (source domain) is used to train an automated machine state identification classifier that can identify the operating states of an eccentric rotating mass vibration motor (target domain). Initial deployment of this method on target-domain motorized devices resulted in a machine status monitoring accuracy of at least 81.6% and a maximum training accuracy of almost 99% on known data of the source domain and 91.49% for unseen data in the target domain within an acceptable time frame. The performance of the proposed method is also comparable across platforms ranging from resource-constrained edge to a resource-rich cloud. This approach facilitates the use of noisy or uncalibrated sensor data in data-driven machine status monitoring tasks, therefore allowing for the development of reusable, low-cost monitoring systems that require meagre developmental effort, resulting in accelerated deployment times.

Empirical Mode Decomposition and Hilbert Spectrum for Abnormality Detection in Normal and Abnormal Walking Transitions.

Sensor-based human activity recognition (HAR) is a method for observing a person's activity in an environment. With this method, it is possible to monitor remotely. HAR can analyze a person's gait, whether normal or abnormal. Some of its applications may use several sensors mounted on the body, but this method tends to be complex and inconvenient. One alternative to wearable sensors is using video. One of the most commonly used HAR platforms is PoseNET. PoseNET is a sophisticated platform that can detect the skeleton and joints of the body, which are then known as joints. However, a method is still needed to process the raw data from PoseNET to detect subject activity. Therefore, this research proposes a way to detect abnormalities in gait using empirical mode decomposition and the Hilbert spectrum and transforming keys-joints, and skeletons from vision-based pose detection into the angular displacement of walking gait patterns (signals). Joint change information is extracted using the Hilbert Huang Transform to study how the subject behaves in the turning position. Furthermore, it is determined whether the transition goes from normal to abnormal subjects by calculating the energy in the time-frequency domain signal. The test results show that during the transition period, the energy of the gait signal tends to be higher than during the walking period.

A multi-source information fusion method for tool life prediction based on CNN-SVM

For milling tool life prediction and health management, accurate extraction and dimensionality reduction of its tool wear features are the key to reduce prediction errors. In this paper, we adopt multi-source information fusion technology to extract and fuse the features of cutting vibration signal, cutting force signal and acoustic emission signal in time domain, frequency domain and time-frequency domain, and downscale the sample features by Pearson correlation coefficient to construct a sample data set; then we propose a tool life prediction model based on CNN-SVM optimized by genetic algorithm (GA), which uses CNN convolutional neural network as the feature learner and SVM support vector machine as the trainer for regression prediction. The results show that the improved model in this paper can effectively predict the tool life with better generalization ability, faster network fitting, and 99.85% prediction accuracy. And compared with the BP model, CNN model, SVM model and CNN-SVM model, the performance of the coefficient of determination R2 metric improved by 4.88%, 2.96%, 2.53% and 1.34%, respectively.

A non-contacting leak fault diagnosis method for subsea Christmas tree valve based on deep neural network with skip connections

Due to the complex and diverse underwater environment, underwater acoustic transmission is easily disturbed, absorbing many noise components. To deal with the complex noise components is a challenge with the traditional signal processing method. Moreover, it is hard to diagnose the valve's leakage and the leakage degree in the subsea Christmas tree using underwater acoustic signals. This paper proposes a new non-contacting fault diagnosis method based on a deep neural network (DNN) with skip connections. First, we obtain the required time-domain data using acoustic sensors (a digital hydrophone) in experiments. Second, the time-domain data is preprocessed and is converted by the short-time Fourier transform into time-frequency domain signals. The time-frequency domain signals are then input into a constructed DNN model to minimize the noise components in the signal. After that, the denoised data are applied to a convolutional neural network for fault diagnosis. Finally, an underwater acoustic experiment is designed and performed to validate the proposed method's effectiveness. The experimental results demonstrate that our proposed method can effectively diagnose the leakage fault of the valve, and the diagnosis accuracy reaches 98.89%.

Bearing fault diagnosis model using improved Bayesian information criterion-based variational modal decomposition and IGA-SVM

In rotating machinery, the bearing vibration signal is easily covered by the vibration signal of other components, which makes the fault characteristic signal not obvious in the collected vibration signal. In order to better separate the vibration source from the collected vibration signal, a variational modal decomposition optimized by bayesian information criterion (VMDBIC) is proposed. At the same time, the data set is constructed based on the separated vibration source signals, and the support vector machine (SVM) optimized by improved genetic algorithm (IGA) is used to diagnose the bearing fault. Firstly, the Bayesian information criterion is used to calculate the number of potential vibration sources in the vibration signal. The calculated number of vibration sources is used as the number of modal components of variational modal decomposition (VMD) to decompose vibration signal. Secondly, the modal components obtained by VMD are used to construct the eigenmatrix of vibration signal. And the singular value energy spectrum of 3 bearing types (a total of 24 fault types) is calculated based on the eigenmatrix of vibration signal. Finally, the singular value energy spectrum is used as the input of IGA-SVM model to diagnose the bearing fault. The effectiveness of VMDBIC method is verified from three aspects of the time-frequency domain signal, the index of orthogonality (IO), and the index of energy conservation (IEC). The feasibility of the method proposed in this paper is verified from three aspects: fault identification effect of various types of bearings, load changes, and mutual diagnosis of different bearings.

A brain-computer interface with gamification in the Metaverse

This study contributes to our understanding of the Metaverse by presenting a case study of the implementation of brain-computer interface supported game-based engagement in a Virtual Environment (VE). In VE, individuals can communicate with anyone, anywhere, anytime, without any limits. This situation will increase the barrier-free living standards of disabled people in a more accessible environment. A virtual world of well-being awaits these individuals, primarily through gamified applications thanks to Brain-Computer Interfaces. Virtual environments in the Metaverse can be infinitely large, but the user's movement in a virtual reality (VR) environment is constrained by the natural environment. Locomotion has become a popular motion interface as it allows for full exploration of VE. In this study, the teleport method from locomotion methods was used. To teleport, the user selects the intended location using brain signals before being instantly transported to that location. Brain signals are decomposed into alpha, beta, and gamma bands. The features of each band signal in Time, frequency, and time-frequency domains are extracted. In this proposed method, the highest performance of binary classification was obtained in the frequency domain and the Alpha band. Signals in the alpha band were tested in the domains Time, Frequency, and Time-Frequency. Teleport operations are faster with Time and more stable with the frequency domain. However, the Hilbert-Huang Transform (HHT) method used in the Time-Frequency domain could not respond adequately to real-time applications. All these analyses were experienced in the Erzurum Virtual Tour case study, which was prepared to promote cultural heritage with the gamification method.

Temporal Convolutional Network with Wavelet Transform for Fall Detection

Fall detection is a challenging task for human activity recognition but is meaningful in health monitoring. However, for sensor-based fall prediction problems, using recurrent architectures such as recurrent neural network models to extract temporal features sometimes could not accurately capture global information. Therefore, an improved WTCN model is proposed in this research, in which the temporal convolutional network is combined with the wavelet transform. Firstly, we use the wavelet transform to process the one-dimensional time-domain signal into a two-dimensional time-frequency domain signal. This method helps us to process the raw signal data efficiently. Secondly, we design a temporal convolutional network model with ultralong memory referring to relevant convolutional architectures. It avoids the gradient disappearance and explosion problem usefully. In addition, this paper also conducts experiments comparing our WTCN model with typical recurrent architectures such as the long short-term memory network in conjunction with three datasets, UniMiB SHAR, SisFall, and UMAFall. The results show that WTCN outperforms other traditional methods, the accuracy of the proposed algorithm is up to 99.53%, and human fall behavior can be effectively recognized in real time.

Partial Discharge (PD) Signal Detection and Isolation on High Voltage Equipment Using Improved Complete EEMD Method

Electricity has a crucial function in contemporary civilization. The power grid must be stable to ensure the efficiency and dependability of electrical equipment. This implies that the high-voltage equipment at the substation must be reliably operated. As a result, the appropriate and dependable use of systems to monitor the operating status of high-voltage electrical equipment has recently gained attention. Partial discharge (PD) analysis is one of the most promising solutions for monitoring and diagnosing potential problems in insulation systems. Noise is a major challenge in diagnosing and detecting defects when using this measurement. This study aims to denoise PD signals using a data decomposition method, improved complete ensemble empirical mode decomposition with adaptive noise algorithm, combined with statistical significance test to increase noise reduction efficiency and to derive and visualize the Hilbert spectrum of the input signal in time-frequency domain after filtering the noise. In the PD signal analysis, both artificial and experimental signals were used as input signals in the decomposition method. For these signals, this study has yielded significant improvement in the denoising and the PD detecting process indicated by statistical measures. Thus, the signal decomposition by using the proposed method is proven to be a useful tool for diagnosing the PD on high voltage equipment.

Enhanced seeded region growing algorithm and its application in signal decomposition

The components in an engineering signal may overlap in the frequency domain, which causes great inconvenience and even errors in traditional one-dimensional signal processing methods. To identify and extract components in the signal from the time-frequency domain, this paper proposes a time-frequency domain signal decomposition method based on enhanced seeded region growing (ESRG). This paper extends the seeded region growing method to adaptive recognition of target regions in the time-frequency representation and divides the time-frequency representation into several time-frequency components. ESRG can reduce the influence of noise, expand the self-adaptability of time-frequency decomposition and avoid errors caused by manually setting the growth point and threshold. The results of the numerical verification and decomposition of bat echo signals show that this method can accurately decompose multi-component non-stationary signals. The decomposition results of rotor fault and rolling bearing fault signals show that this method can be successfully applied to fault diagnosis.

Identifying microseismic events using a dual-channel CNN with wavelet packets decomposition coefficients

Microseismic monitoring is a widely used technique in coal mine safe production. Microseismic events classification is one of the most important steps in microseismic monitoring. CNN has been proven to be the potential tool to identify microseismic events automatically. When the noise is serious and the signal is weak, the recognition ability of CNN is limited. Some scholars use denoising methods, such as wavelet transform, to enhance the signal-to-noise ratio (SNR) of the input data. However, the effective signal will be broken if the incorrect threshold parameters are set in this wavelet transform-based denoising method. In this paper, we intend to make the data volume of the time-frequency graph obtained by WPT as the input of CNN. But it may lead to a dimensional disaster if taking the time-frequency domain signals as input directly. To utilize the effective information extraction and reduce the data dimension, we propose a dual-channel CNN model by combining time domain information and wavelet packet decomposition coefficients (T-WPD CNN). The wavelet packet decomposition coefficients highlight the characteristics of the signal and suppress the characteristics of noise. In addition, the coefficients have the same dimension as the original signal. These advantages are useful for enhancing the performance of CNN. Two field datasets are used to test the network. One from Australia and another from Jiayang coal mine, Sichuan, China. The feature map of the convolutional layer with different inputs are shown to illustrate the influence effect of raw time domain signal and WPD coefficient on the classification performance. The final results show that the T-WPD CNN is superior to the traditional CNN method in accuracy and robustness.