Signal Feature Extraction Research Articles

Fine-Grained Recognition of Mixed Signals with Geometry Coordinate Attention.

With the advancement of technology, signal modulation types are becoming increasingly diverse and complex. The phenomenon of signal time-frequency overlap during transmission poses significant challenges for the classification and recognition of mixed signals, including poor recognition capabilities and low generality. This paper presents a recognition model for the fine-grained analysis of mixed signal characteristics, proposing a Geometry Coordinate Attention mechanism and introducing a low-rank bilinear pooling module to more effectively extract signal features for classification. The model employs a residual neural network as its backbone architecture and utilizes the Geometry Coordinate Attention mechanism for time-frequency weighted analysis based on information geometry theory. This analysis targets multiple-scale features within the architecture, producing time-frequency weighted features of the signal. These weighted features are further analyzed through a low-rank bilinear pooling module, combined with the backbone features, to achieve fine-grained feature fusion. This results in a fused feature vector for mixed signal classification. Experiments were conducted on a simulated dataset comprising 39,600 mixed-signal time-frequency plots. The model was benchmarked against a baseline using a residual neural network. The experimental outcomes demonstrated an improvement of 9% in the exact match ratio and 5% in the Hamming score. These results indicate that the proposed model significantly enhances the recognition capability and generalizability of mixed signal classification.

Hydroelectric Unit Vibration Signal Feature Extraction Based on IMF Energy Moment and SDAE

Hydroelectric Unit Vibration Signal Feature Extraction Based on IMF Energy Moment and SDAE

Numerical Simulation and Application of Vortex Field Monitoring near Islands in Straits

This study addresses the issue of vortex current fields near the islands and reefs in China’s straits, which pose significant challenges to engineering construction and navigation safety in the surrounding waters. To monitor these vortex fields, the study proposes an innovative method utilizing acoustic signals. The study utilizes the numerical simulation and phase feature extraction of acoustic signals in the vortex current field, based on ray acoustic theory and time-reversal mirror technology. The study successfully monitored the central position of the vortex core and characteristic radius of the vortex current field near Barley Straw Reef using HYCOM data for the first time. Furthermore, the performance of the method was analyzed under different acoustic phase perturbations and signal-to-noise ratios. The numerical simulation results demonstrate that the acoustic method is effective in monitoring near-shore vortex fields, and the time-reversal mirror technique is useful in extracting phase difference information from acoustic signals generated by vortex currents.

Feature Extraction Methods for Underwater Acoustic Target Recognition of Divers.

The extraction of typical features of underwater target signals and excellent recognition algorithms are the keys to achieving underwater acoustic target recognition of divers. This paper proposes a feature extraction method for diver signals: frequency-domain multi-sub-band energy (FMSE), aiming to achieve accurate recognition of diver underwater acoustic targets by passive sonar. The impact of the presence or absence of targets, different numbers of targets, different signal-to-noise ratios, and different detection distances on this method was studied based on experimental data under different conditions, such as water pools and lakes. It was found that the FMSE method has the best robustness and performance compared with two other signal feature extraction methods: mel frequency cepstral coefficient filtering and gammatone frequency cepstral coefficient filtering. Combined with the commonly used recognition algorithm of support vector machines, the FMSE method can achieve a comprehensive recognition accuracy of over 94% for frogman underwater acoustic targets. This indicates that the FMSE method is suitable for underwater acoustic recognition of diver targets.

A VMD-BP Model to Predict Laser Welding Keyhole-Induced Pore Defect in Al Butt-Lap Joint.

The detection of keyhole-induced pore positions is a critical procedure for assessing laser welding quality. Considering the detection error due to pore migration and noise interference, this research proposes a regional prediction model based on the time-frequency-domain features of the laser plume. The original plume signal was separated into several signal segments to construct the morphological sequences. To suppress the mode mixing caused by environmental noise, variational modal decomposition (VMD) was utilized to process the signals. The time-frequency features extracted from the decomposed signals were acquired as the input of a backpropagation (BP) neural network to predict the pore locations. To reduce the prediction error caused by pore migration, the effect of the length of the signal segments on the prediction accuracy was investigated. The results show that the optimal signal segment length was 0.4 mm, with an accuracy of 97.77%. The 0.2 mm signal segments failed to eliminate the negative effects of pore migration. The signal segments over 0.4 mm resulted in prediction errors of small and dense pores. This work provides more guidance for optimizing the feature extraction of welding signals to improve the accuracy of welding defect identification.

Underwater Acoustic Target Classification Using Convolutional Neural Network Combined with Continuous Wavelet Transform

Underwater acoustic target (UAT) classification is a critical task in submarine warfare because acoustic waves are the reliable information source that allows sonar operators and commanders to understand the surrounding situation in the operational area. To solve the problem of UAT classification, this paper proposes a convolutional neural network combined with continuous wavelet transform for improving the accuracy of UAT classification. The classification focuses on types of noise emitted from the propellers of different ships. Signal processing methods such as Short Time Fourier Transform, Continuous Wavelets Transform, and CNN are executed to extract signal features that provide information for classifier. Simulation results are achieved with an accuracy of 99.64\% when using the Continuous Wavelets Transform method combined with the convolutional neural network, which is higher than other traditional methods.

Research on a new method of islanding detection based on lifting wavelet and neural network

Abstract At present, the commonly used active and passive islanding detection methods have their own shortcomings, and the islanding detection effect is difficult to meet the requirements. Therefore, a new detection method based on wavelet signal processing and artificial intelligence to identify islanding is proposed in this paper. In this method, the feature quantities required for islanding detection are obtained by wavelet transform and signal processing, and then the feature quantities are identified by neural network to determine whether the islanding is generated in distributed generation system. Wavelet transformation has a strong ability of signal feature extraction, while the neural network has strong learning and identification abilities, the combination of the both is beneficial to improve the success rate of islanding detection. Simulation verification shows that the new islanding detection method proposed in this paper can detect islanding quickly and accurately, and the performance of islanding detection has been significantly improved.

A diagnosis method for loss of circulation based on transient-pressure wave analysis and particle swarm optimization

Lost circulation during operations poses a significant threat to production processes. In the search for an effective detection method, an impulse-response detection method of lost circulation is introduced. This method involves generating transient pressure waves at the wellhead and analyzing their time-frequency domain characteristics to pinpoint location for lost circulation within the wellbore annulus system. Utilizing the data processing capabilities of machine learning models, this study proposes an integrated model for signal feature classification and diagnosis model for lost circulation. Drawing from extensive experimental data, this model integrates laboratory experiments, signal analysis, and machine learning algorithms. Data preprocessing, including wavelet variation and denoising, precedes the application of an enhanced adaptive noise complete ensemble empirical modal decomposition with adapted noise (ICEEMDAN) alongside energy and sample entropy analysis for feature extraction. By establishing a mapping relationship between signal features and lost circulation changes, we develop an improved backpropagation neural network (IBP) model using the particle swarm optimization (PSO) algorithm for diagnosis (PSO-IBP). Comparative analysis of five models reveals compelling results: ① PSO-IBP achieves an average accuracy of 97.60%, with a standard deviation of 0.356; ② diagnosis accuracy for every lost circulation scenario exceeds 92%, outperforming other models in precision, recall, and F-Score; ③ even with limited training data, PSO-IBP maintains 84% accuracy, demonstrating superior performance. Further analysis highlights the efficacy of PSO-IBP, especially when leveraging ICEEMDAN for signal feature extraction, in accurately diagnosing lost circulation.

Dynamic multiscale pressure fluctuation features extraction of mixed-flow pump as turbine (PAT) and flow state recognition of the outlet passage using variational mode decomposition and refined composite variable-step multiscale multimapping dispersion entropy

The feature extraction of pressure fluctuation signal (PFS) and flow state recognition of outlet passage have crucial engineering significance to guarantee the safe and reliable operation in mixed-flow pump as turbine (PAT). To accurately extract the dynamic multiscale features of PFS, a method based on variational mode decomposition (VMD) and refined composite variable-step multiscale multimapping dispersion entropy (RCVMMDE) is proposed. By applying VMD to PFS, intrinsic mode functions (IMFs) are obtained, and the RCVMMDE values for each IMF is then calculated. Model parameters based on the RCVMMDE indicator are then established and used as feature vectors for flow state recognition. Using the PFS at the outlet passage inlet as an example, this method extracts dynamic multiscale feature information of the outlet passage, which is validated through experimental and numerical simulations. The results show that this method achieves high accuracy, providing well-defined feature vectors and effectively capturing the dynamic multiscale features of the PAT and turbine systems.

Series alternating current arc fault detection method based on relative position matrix and deep convolutional neural network

The detection of series arc faults is crucial due to the potential safety hazards caused by cable aging and breakage. This paper proposes an arc fault detection framework that utilizes the relative position matrix and deep convolutional network to accurately detect alternating current series arc faults. In this framework, the time series data is first converted into a relative position matrix and then visualized as a two-dimension image. This enables the characterization of singular values and flat shoulders in the arc current, making it easier to extract significant features for the models. Next, a Residual Network with hybrid attention mechanisms model is constructed and the images are fed into the model to determine whether an arc fault has occurred in the current signal. Finally, the data is collected using the experimental platform, and the relative position matrix is constructed with different dimensionality reduction factor. The appropriate dimensionality reduction factor is selected through experimental comparisons. In addition, the effectiveness of the proposed method is verified by measured data, and the proposed model is compared with four other commonly used network models, which demonstrates the superiority of the proposed model in terms of detection performance. The proposed method achieves a detection accuracy of 98.74%.

DSWIN: Automated hunger detection model based on hand-crafted decomposed shifted windows architecture using EEG signals

Hunger is a physiological state that arises from complex interactions of multiple factors, including higher brain center control. We purposed to develop an accurate and efficient machine-learning model for the automated detection of hunger using EEG signals. We prospectively acquired 14-channel EEG (sampling frequency 128 Hz) from 43 and 48 fasted and post-prandial healthy subjects (hungry vs. control groups, respectively) using the EMOTIV EPOC+ mobile brain cap system. To augment the hunger response, fasted subjects were also shown video images of food during EEG recording. The EEG signals were divided into 15-second segments. 877 and 852 participants/subjects were in the hungry and control groups. We created a novel handcrafted architecture—decomposed shifted window (DSWIN)—that combined swin patch division with tunable Q-factor wavelet transform-based signal decomposition for multilevel feature extraction of EEG signals. Textural and statistical features were extracted from the multiple patches and decomposed signals using a novel penta pattern-based extractor and statistical moments, respectively, and then merged. Iterative neighborhood component analysis (INCA) and iterative ReliefF (IRF) were applied. Twenty-eight selected feature vectors were generated, which were then fed to a shallow k-nearest neighbors (kNN) classifier to calculate channel-wise prediction vectors. From the 28 channel-wise prediction vectors, another 26 modes of function-based voted results were calculated using iterative hard majority voting, and the best overall model result was selected using a greedy algorithm. Our model attained 99.54% and 82.71% binary classification accuracies of hungry status vs. control using 10-fold and leave-one-subject-out cross-validations, respectively.

Dam Deformation Prediction Model Based on Multi-Scale Adaptive Kernel Ensemble

Aiming at the noise and nonlinear characteristics existing in the deformation monitoring data of concrete dams, this paper proposes a dam deformation prediction model based on a multi-scale adaptive kernel ensemble. The model incorporates Gaussian white noise as a random factor and uses the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) method to decompose the data set finely. Each modal component is evaluated by sample entropy (SE) analysis so that the data set can be reconstructed according to the sample entropy value to retain key information. In addition, the model uses partial autocorrelation function (PACF) to determine the correlation between intrinsic modal function (IMF) and historical data. Then, the global search whale optimization algorithm (GSWOA) is used to accurately determine the parameters of kernel extreme learning machine (KELM), which forms the basis of the dam deformation prediction model based on multi-scale adaptive kernel function. The case analysis shows that CEEMDAN-SE-PACF can effectively extract signal features and identify significant components and trends so as to better understand the internal deformation trend of the dam. In terms of algorithm optimization, compared with the WOA algorithm and other algorithms, the results of the GSWOA algorithm are significantly better than other algorithms and have the optimal convergence. In terms of prediction performance, CEEMDAN-SE-PACF-GSWOA-KELM is superior to the CEEMDAN-WOA-KELM, GSWOA-KELM, CEEMDAN-KELM, and KELM models, showing higher accuracy and stronger stability. This improvement is manifested in the decrease of root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE) and the improvement of the R square (R2) value close to 1. These research results provide a new method for dam safety monitoring and evaluation.

Application of machine learning for signal recognition in distributed fibre optic acoustic sensing technology

AbstractCoherent Rayleigh scattering‐based distributed fibre optic sensing technology enables real‐time acquisition of vibration and acoustic information along the optical fibres. However, the complexity of monitoring environments often leads to false alarms and missed detections during the process of information source identification with distributed acoustic sensing (DAS). Therefore, it becomes crucial to effectively extract meaningful signal features and perform accurate pattern recognition in the presence of external noise disturbance. The authors provide a comprehensive review of signal feature extraction and pattern recognition techniques applied in DAS technology. After introducing the fundamentals of DAS, specific applications are considered, and the following techniques have been analysed and compared: feature extraction algorithms based on wavelet decomposition, feature extraction schemes utilising other decomposition models, traditional recognition classifiers, and neural network‐based recognition classifiers using deep learning. The advantages and limitations of each scheme are discussed, along with their potential applications in various scenarios. The aim is to provide insights into the latest technologies in signal processing and pattern recognition for DAS, fostering further advancements in this field.

An intelligent biosensing platform using phase space reconstruction‐assisted convolutional neural network for drug‐induced cardiotoxicity assessment

AbstractDrug‐induced cardiotoxicity often leads to patient deaths and drug recalls. Interdigitated electrode (IDE)‐based cellular impedance detection instrument has been integrated with intelligent algorithms to screen cardiotoxicity based on cardiomyocytes. These intelligent biosensing systems generally employ traditional machine learning methods to assess drug‐induced cardiotoxicity by analyzing cardiomyocytes mechanical beating signals. However, modern deep learning methods with robustness and flexibility have not been integrated in IDE‐based platform to screen cardiotoxicity. Here, for the first time, we implemented deep convolutional neural network (CNN) to analyze cardiomyocytes mechanical beating signals for cardiotoxicity assessment. This method can eliminate the feature engineering procedures, such as manual design and extraction of signal features required by traditional machine learning methods. To facilitate two‐dimensional (2‐D) CNN analysis, we utilized phase space reconstruction to convert one‐dimensional beating signals into 2‐D image representations as well as take into account nonlinear dynamic information. The phase space reconstruction‐assisted convolutional neural network (PSRCNN) is capable of accurately categorizing drug‐induced cardiotoxicities and predicting cardiotoxicity levels. It obtains accuracies ranging from 0.82 to 0.99 when recognizing the cardiotoxicity of newly developed drugs, which are represented by drugs that are not used during model training. Furthermore, we explore in depth to compare the performance of CNN with other traditional machine learning methods. The PSRCNN method can achieve higher accuracies when compared to other methods. The PSRCNN‐based biosensing platform is highly potential in improving the efficiency and accuracy of high‐throughput screening of newly developed drugs for cardiotoxicity during drug discovery.

Adaptive feature extraction of underwater target signal based on mathematical morphology feature enhancement

This paper proposes an adaptive feature extraction method based on mathematical morphology enhancement to extract effective and stable features under strong ocean noise. Firstly, the traditional mathematical morphology method is improved and a new mathematical morphology filtering method is proposed for feature enhancement of underwater target signals. Secondly, the artificial fish swarm algorithm (AFSA) is improved using weighted power spectral kurtosis (WPSK) to achieve parameter adaptivity. Five measured underwater target signals are used to validate the extracted method, and cosine similarity (CS), signal to noise ratio (SNR), and refined composite multiscale fluctuation dispersion entropy (RCMFDE) are used as evaluation metrics to assess the results and verify the effectiveness of the proposed method. Compared with average filtering (AVGF) and closing opening filtering (COF), the proposed method shows better capability in adaptive feature extraction. Therefore, the proposed mathematical morphological filtering can enhance underwater target features and is very important for adaptive feature extraction of underwater targets.

Impact of potential function asymmetry on the performance of a novel stochastic resonance system

Due to output saturation and parameter coupling, the traditional classical bistable stochastic resonance (CBSR) system is often unable to effectively handle the extraction of weak signal features encountered in many engineering applications. To overcome this limitation, an asymmetric decoupled unsaturated bistable stochastic resonance (ADUBSR) system is proposed in this paper. Subsequently, the impact of barrier asymmetry on system output has been analyzed by taking signal-to-noise ratio (SNR), peak signal-to-noise ratio (PSNR), and the optimal noise required for obtaining PSNR as evaluation indicators. The results indicate that: (1) The output SNR is effectively enhanced by barrier-width asymmetry only when the left barrier is narrower than the right. (2) Barrier-height asymmetry fails to effectively improve the output SNR. (3) In the case of barrier-height-width asymmetry, the barrier width asymmetry predominantly enhances the output SNR, resulting in the best performance. (4) For all three asymmetric structures, regardless of noise type, the optimal noise intensity depends greatly on barrier height and width values, and is less influenced by asymmetry. Finally, the proposed system is used for engineering data, and the results show that the signal processed by the ADUBSR system achieves a higher output SNR and the amplitude at the fault frequency is 10 times greater than that of the signal processed by the CBSR system.

Processing and analysis of friction vibration signal of diamond-like carbon film microtextured dry gas seal rings

PurposeThis paper aims to analyze the characteristics of friction vibration signals and identify the vibration excitation source at the start and stop stage of microtextured end face of dry gas seals.Design/methodology/approachThe friction pair consists of a diamond-like carbon (DLC) film microtextured seal ring and a spiral groove seal ring. Friction vibration signal feature extraction method based on harmonic wavelet packet and spectrum analysis was proposed. Signals were collected using acceleration sensor, acquisition card and LabVIEW software. Vibration acceleration signal was decomposed into 32 frequency bands using MATLAB wavelet packet transformation. The 32nd band coefficient was extracted for reconstruction, time-domain and spectral waveforms were obtained and spectra before/after denoising were compared.FindingsThe end face of the DLC film microtextured seal ring generates a good dynamic pressure effect, and the friction and vibration reduction effects are obvious. The harmonic wavelet packet can decompose the vibration signal conveniently and precisely. In the case of this experiment, the frequency of vibration of the seal ring is 7500 HZ.Originality/valueThe results show that the method is effective for the processing of friction vibration signal and the identification of vibration excitation source. The findings will provide ideas for the frictional vibration signal processing and basis for further research in the field of tribology of dry gas seal ring.Peer reviewThe peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-03-2024-0084/

A Hybrid Model for Detecting Intrusions on Network Logs

The presence of malicious traffic presents a substantial risk to network systems and the integrity of confidential information. Organisations may enhance their protection against threats and mitigate the possible impact of malicious traffic on their networks by maintaining vigilance, deploying comprehensive security measures, and cultivating a cybersecurity-aware culture. The purpose of this study is to propose a theoretical framework for identifying and analysing potentially harmful network traffic within a network system. In order to identify and classify various types of malicious network traffic in a multi-class setting, we employed a dataset consisting of nine distinct categories of network system attacks. In order to optimise the performance of the model, an exploratory data analysis is conducted on the dataset. Exploratory data analysis (EDA) was employed to assess various aspects like the presence of missing values, correlation among characteristics, data imbalance, and identification of significant features. The findings derived from the exploratory data analysis indicate that the dataset exhibits an imbalance, which, if left unaddressed, may result in overfitting. The data imbalance was addressed with the implementation of the RandomOverSampling approach in Python, which involved executing random oversampling. Following the resolution of the data imbalance, a random forest classifier was employed to extract significant features from the dataset. In this study, a total of ten characteristics were extracted based on the ranking provided by the random forest model. The features that were extracted were utilised in the training process of the suggested model, which aims to identify and detect malicious activity within a network system. The findings of the model indicate a much improved level of accuracy in identifying malicious traffic within a network system, with an accuracy rate of 99.99%. Furthermore, the precision, recall, and F1-score metrics also demonstrate a consistent accuracy rate of 99.99%.

Iterative feature mode decomposition: a novel adaptive denoising method for mechanical fault diagnosis

Remaining useful life prediction of rolling bearings highly relies on feature extraction of signals. The use of denoising algorithms helps to better eliminate noise and extract features, thereby constructing health indicators to predict remaining useful life. This paper proposes a novel adaptive denoising method based on iterative feature mode decomposition (IFMD) to accurately and efficiently extract fault features. The feature mode decomposition (FMD) employs correlation kurtosis (CK) as the objective function for iterative filter bank updates, enabling rapid identification of fault features. To achieve IFMD, the sparrow search algorithm combines sine-cosine algorithm and cauchy variation (SCSSA) to optimize two key parameters in FMD. During the continuous iteration process of the SCSSA algorithm, filter length and number of modes were determined. IFMD does not require empirical setting of initial parameters. During iterative process, the signal is accurately decomposed and the noise is eliminated. Compared with other optimization algorithms, SCSSA has obvious advantages in iterative rate and global optimization. The envelope spectrum feature energy ratio (ES-FER) is used to select decomposed modes, and the mode with the largest ES-FER is chosen as the optimal mode. Bearing fault diagnosis is realized by envelope spectrum analysis of the optimal mode. The numerical simulations and experimental verifications both validate the effectiveness and superiority of the proposed IFMD in mechanical fault diagnosis.

Inter-turn Short Circuit Fault Diagnosis and Severity Estimation for Wind Turbine Generators

While preventive maintenance is crucial in wind turbine operation, conventional condition monitoring systems face limitations in terms of cost and complexity when compared to innovative signal processing techniques and artificial intelligence. In this paper, a cascading deep learning framework is proposed for the monitoring of generator winding conditions, specifically to promptly detect and identify inter-turn short circuit faults and estimate their severity in real time. This framework encompasses the processing of high-resolution current signal samples, coupled with the extraction of current signal features in both time and frequency domains, achieved through discrete wavelet transform. By leveraging long short-term memory recurrent neural networks, our aim is to establish a cost-efficient and reliable condition monitoring system for wind turbine generators. Numeral experiments show an over 97% accuracy for fault diagnosis and severity estimation. More specifically, with the intrinsic feature provided by wavelet transform, the faults can be 100% identified by the diagnosis model.