Time-frequency Distribution Images Research Articles

Localization method of subsynchronous oscillation source based on high-resolution time-frequency distribution image and CNN

The penetration of new energy sources such as wind power is increasing, which consequently increases the occurrence rate of subsynchronous oscillation events. However, existing subsynchronous oscillation source-identification methods primarily analyze fixed-mode oscillations and rarely consider time-varying features, such as frequency drift, caused by the random volatility of wind farms when oscillations occur. This paper proposes a subsynchronous oscillation source-localization method that involves an enhanced short-time Fourier transform and a convolutional neural network (CNN). First, an enhanced STFT is performed to secure high-resolution time-frequency distribution (TFD) images from the measured data of the generation unit ports. Next, these TFD images are amalgamated to form a subsynchronous oscillation feature map that serves as input to the CNN to train the localization model. Ultimately, the trained CNN model realizes the online localization of subsynchronous oscillation sources. The effectiveness and accuracy of the proposed method are validated via multimachine system models simulating forced and natural oscillation events using the Power Systems Computer Aided Design platform. Test results show that the proposed method can localize subsynchronous oscillation sources online while considering unpredictable fluctuations in wind farms, thus providing a foundation for oscillation suppression in practical engineering scenarios.

Blind Detection of Broadband Signal Based on Weighted Bi-Directional Feature Pyramid Network.

With the development of wireless technology, signals propagating in space are easy to mix, so blind detection of communication signals has become a very practical and challenging problem. In this paper, we propose a blind detection method for broadband signals based on a weighted bi-directional feature pyramid network (BiFPN). The method can quickly perform detection and automatic modulation identification (AMC) on time-domain aliased signals in broadband data. Firstly, the method performs a time-frequency analysis on the received signals and extracts the normalized time-frequency images and the corresponding labels by short-time Fourier transform (STFT). Secondly, we build a target detection model based on YOLOv5 for time-domain mixed signals in broadband data and learn the features of the time-frequency distribution image dataset of broadband signals, which achieves the purpose of training the model. The main improvements of the algorithm are as follows: (1) a weighted bi-directional feature pyramid network is used to achieve a simple and fast multi-scale feature fusion approach to improve the detection probability; (2) the Efficient-Intersection over Union (EIOU) loss function is introduced to achieve high accuracy signal detection in a low Signal-Noise Ratio (SNR) environment. Finally, the time-frequency images are detected by an improved deep network model to complete the blind detection of time-domain mixed signals. The simulation results show that the method can effectively detect the continuous and burst signals in the broadband communication signal data and identify their modulation types.

A deep learning approach for rapid detection of soil liquefaction using time–frequency images

Soil liquefaction – a common hazard during earthquakes – can cause severe damage to infrastructure systems. The excessive pore water pressure induced by underground shaking can result in a sudden loss of soil stiffness and strength, which renders the soil–water mixture to behave temporarily as “liquid”. Identifying liquefaction events and extracting their time of initiation and duration from recorded data are critical capabilities for validating and improving predictive models, and ultimately, effectively mitigating this hazard. The current approach is to conduct on-site investigations or simplified statistical analyses, which are generally time-consuming and bear high uncertainties, except for the most obvious cases. In this study, we utilize a convolutional neural network (CNN), a deep learning method, to predict both the occurrence of a liquefaction event and its triggering time (if it occurred). For this purpose, first, more than one million high-fidelity nonlinear site response simulations with different soil profiles and ground motions are performed. The surface acceleration time-histories are then converted into time–frequency distribution images, which are fed into a CNN as the training inputs. The trained CNN exhibits high prediction accuracy for both training and validation sets, another suite of independent site response simulations, as well as real-life earthquake data.

Application of Augmented Spread Spectrum Time Domain Reflectometry for Detection and Localization of Soft Faults on a Coaxial Cable

Aviation cables are an essential part of aircraft, which transmits signals and power. Aviation cable faults often occur due to drastic changes in temperature and humidity and the effects of vibration. Obvious cable faults, such as open circuits and short circuits, can be detected during maintenance. However, some soft faults like the loose connection of the connector, damage of the insulation layer, and improper cable bending radius cannot be detected. Spread spectrum time domain reflectometry (SSTDR) is a diagnostic method for cable faults; however, the amplitude of the reflected signal is low when the soft fault occurs. SSTDR is only sensitive to apparent faults, but it is insufficient to detect soft faults. This article proposes an aviation cable fault detection and location method, augmented spread spectrum time domain reflectometry (ASSTDR), to detect soft faults effectively. First, the SSTDR results are decomposed into some intrinsic mode functions (IMFs) using the improved variational mode decomposition and selects the IMF with the highest kurtosis to analyze. Second, short-time Fourier transform (STFT) is applied to deal with the selected IMF to enhance the amplitude characteristics. Then, reassign spectrogram is computed to improve the resolution of STFT time-frequency distribution images to locate the location of soft faults. Finally, the effectiveness of ASSTDR is verified by experiments with different fault types and fault degrees. The comparative experiments demonstrate that the method can accurately detect soft faults.

Modulation Recognition of Radar Signals Based on Adaptive Singular Value Reconstruction and Deep Residual Learning.

Automatically recognizing the modulation of radar signals is a necessary survival technique in electronic intelligence systems. In order to avoid the complex process of the feature extracting and realize the intelligent modulation recognition of various radar signals under low signal-to-noise ratios (SNRs), this paper proposes a method based on intrapulse signatures of radar signals using adaptive singular value reconstruction (ASVR) and deep residual learning. Firstly, the time-frequency spectrums of radar signals under low SNRs are improved after ASVR denoising processing. Secondly, a series of image processing techniques, including binarizing and morphologic filtering, are applied to suppress the background noise in the time-frequency distribution images (TFDIs). Thirdly, the training process of the residual network is achieved using TFDIs, and classification under various conditions is realized using the new-trained network. Simulation results show that, for eight kinds of modulation signals, the proposed approach still achieves an overall probability of successful recognition of 94.1% when the SNR is only −8 dB. Outstanding performance proves the superiority and robustness of the proposed method.

Frequency Measurement of Dynamic Stress in Polarization Maintaining Fibers

Up to now, the distributed polarization coupling (DPC) measurement system based on white light interferometry (WLI) has been mainly studied for static and quasi-static measurement. In this paper, we developed the DPC system to measure the instantaneous frequency of dynamic stress. First, we set up a model of transversal stress on the panda polarization maintaining fiber (PMF) and simulated the interferograms of dynamic coupling points when the PMF is subjected to stress of cosine wave, sawtooth wave and square wave. Then, we did groups of experiments to measure the interferograms of dynamic coupling points, including cosinoidal coupling point, sawtooth wave coupling point, square wave coupling point, and two cosinoidal coupling points. By applying wavelet transform to the envelops of interferograms, we can obtain the time-frequency distribution images and demodulate the instantaneous frequency of external stress. The absolute error of experiments is less than 1.84 Hz. Finally, we deduced the upper and lower limits for frequency measurement and discussed the methods of enlarging the frequency range in detail. The method introduced in this paper has potential application in fiber sensing of dynamic stress.

A Doppler aliasing free micro-motion parameter estimation method in the terahertz band

Micro-Doppler, induced by micro-motion of targets, is an important characteristic for target recognition once extracted via parameter estimation. However, micro-Doppler is usually too significant to result in aliasing in the terahertz band. According to this problem, a Doppler aliasing free micro-motion parameter estimation method based on the modulo Generalized Hough transform is proposed in this paper. Its basic idea is to search and match the parameters of aliasing micro-Doppler curves in the time-frequency distribution image. The high estimation precision and excellent noise immunity of this method are verified through simulations based on CST data and experiments based on a 0.33 THz radar system.

LPI Radar Waveform Recognition Based on Time-Frequency Distribution.

In this paper, an automatic radar waveform recognition system in a high noise environment is proposed. Signal waveform recognition techniques are widely applied in the field of cognitive radio, spectrum management and radar applications, etc. We devise a system to classify the modulating signals widely used in low probability of intercept (LPI) radar detection systems. The radar signals are divided into eight types of classifications, including linear frequency modulation (LFM), BPSK (Barker code modulation), Costas codes and polyphase codes (comprising Frank, P1, P2, P3 and P4). The classifier is Elman neural network (ENN), and it is a supervised classification based on features extracted from the system. Through the techniques of image filtering, image opening operation, skeleton extraction, principal component analysis (PCA), image binarization algorithm and Pseudo–Zernike moments, etc., the features are extracted from the Choi–Williams time-frequency distribution (CWD) image of the received data. In order to reduce the redundant features and simplify calculation, the features selection algorithm based on mutual information between classes and features vectors are applied. The superiority of the proposed classification system is demonstrated by the simulations and analysis. Simulation results show that the overall ratio of successful recognition (RSR) is 94.7% at signal-to-noise ratio (SNR) of −2 dB.

An image dimensionality reduction method for rolling bearing fault diagnosis based on singular value decomposition

The fast kurtogram, a faint signal extraction method, has been regarded as an effective approach to detect and characterize faint transient features in vibration signals. However, the fast kurtogram, a band-pass filtering method, which extracts transient signals by optimal frequency band selection and leaves the noise in the selected frequency band unprocessed. Therefore, to overcome the shortcoming of the fast kurtogram method, a method which can wipe off the noise in the whole frequency band is necessary. This paper proposes a novel faint signal extraction method by time–frequency distribution image dimensionality reduction. Since time–frequency distribution image can reveal intrinsic feature of nonstationary signals and can make the weak impulses feature prominent, and besides, the transient impulse feature and the noise component lie in different dimensions, so using the dimensionality reduction method based on singular value decomposition to suppress the background noise in the raw time–frequency distribution image is motivated. A bearing outer race fault signal obtained from a test-to-failure experiment and a bearing inner race fault signal obtained from an experimental motor are employed to demonstrate the enhanced performance of the proposed method in faint signal extraction. The results indicate that the proposed method outperforms the fast kurtogram method and is effective in faint signal extraction.

Rotating Parabolic-Reflector Antenna Target in SAR Data: Model, Characteristics, and Parameter Estimation

Parabolic-reflector antennas (PRAs), usually possessing rotation, are a particular type of targets of potential interest to the synthetic aperture radar (SAR) community. This paper is aimed to investigate PRA’s scattering characteristics and then to extract PRA’s parameters from SAR returns, for supporting image interpretation and target recognition. We at first obtain both closed-form and numeric solutions to PRA’s backscattering by geometrical optics (GO), physical optics, and graphical electromagnetic computation, respectively. Based on the GO solution, a migratory scattering center model is at first presented for representing the movement of the specular point with aspect angle, and then a hybrid model, named the migratory/micromotion scattering center (MMSC) model, is proposed for characterizing a rotating PRA in the SAR geometry, which incorporates PRA’s rotation into its migratory scattering center model. Additionally, we in detail analyze PRA’s radar characteristics on radar cross-section, high-resolution range profiles, time-frequency distribution, and 2D images, which also confirm the models proposed. A maximal likelihood estimator is developed for jointly solving the MMSC model for PRA’s multiple parameters by optimization. By exploiting the aforementioned characteristics, the coarse parameter estimation guarantees convergency upon global minima. The signatures recovered can be favorably utilized for SAR image interpretation and target recognition.

Modernized method of estimating the parameters of LFM signals based on the time-frequency distribution correction and the use of the Hough transformation

The method has been modernized for obtaining the parameter estimation of the fine structure of LFM signals (linear-frequency-modulated signals) at small values of the signal-to-noise ratio. The development of this method was based on the analysis of the signal time-frequency distribution (TFD) and the Hough transform. The specific feature of this method is correction of the time-frequency parameters in the TFD image considered and the use of the principle of detecting the straight line by the Hough transform that allows us to obtain estimates of parameters of LFM signals from the radio signal received within a shorter time interval at small signal-to-noise ratios. The results of simulation modeling demonstrate the capabilities of the method proposed.

Wavelet Packet Images Matching Applied to Noise Faults Diagnosis

Abstract In order to avoid the difficulty of installing vibration sensors and extracting characteristic frequency vectors on the traditional vibration-based abrasion faults diagnosis of main bearing of diesel engine, this paper presents a new approach based on the noise signal of diesel engine and wavelet packet images processing. Based on that, the standard time-frequency distribution images of all fault conditions, including the gap abrasion information of main bearing, can be defined. Correspondingly, a gap abrasion fault diagnosis model of main bearing with images matching is set up. Through comparing the Euclid Distance values between standard fault images and the test image, the model can recognize the gap abrasion condition. The result shows that this method is simple and effective, and makes the best use of fault information.