Search

In the field of neuroscience, a significant challenge lies in extracting essential features from biological signals like Electroencephalography (EEG). Utilized as a non-invasive method, EEG records brain activities through metal electrodes on the scalp. The analysis of EEG data finds applications in various domains, including concealed information tests, aimed at detecting deception. This paper introduces the Spatial Spiking Neural Network, a supervised approach for classifying EEG data collected during concealed information tests. Temporal EEG data undergoes filtration using a Finite Impulse Response (FIR) filter, while Common Spatial Pattern (CSP) is employed to extract spatial components. Binary classification is achieved through an integrate-and-fire neuron model, where the frequency of spike generation determines the classification. Spiking Neural Networks (SNNs) offers advantages in terms of temporal precision, event-driven processing, and low power consumption. Their spike-based communication allows for efficient handling of sparse data and recognition of temporal patterns, contributing to robustness and energy efficiency. The proposed model is applied separately to each subject's EEG data, and the results are compared with traditional classification algorithms. The proposed approach attains a peak accuracy of 90.15%, showcasing superior performance compared to alternative methods. [ABSTRACT FROM AUTHOR]

In shipborne high-frequency surface wave radar (HFSWR), the change of platform speed or heading will cause variations in the extension of the Doppler spectrum for vessel targets located in different directions, subsequently resulting in alterations to the signal-to-noise ratio (SNR). These alterations in vessel target echo present challenges for manual parameter adjustments in traditional constant false alarm rate (CFAR) methods for shipborne HFSWR, thereby hindering the maintenance of stable target detection capabilities. In this paper, a self-attention-convolutional neural network (SA-CNN)-based CFAR detector is proposed, which transforms the detection problem into signal structure classification. First, the extension characteristics of vessel target echoes resulting from changes in speed or heading of the shipborne platform are quantitatively analysed, thereby guiding the selection of an optimal sliding window and constructing input vectors for the neural network. Subsequently, the SA-CNN is designed to efficiently extract the structural features of the signal and accurately predict the probability of target presence. Finally, the Monte Carlo method is used to control the false alarm rate effectively. Simulation and real dataset verification demonstrate that the proposed method exhibits superior detection performance compared to traditional methods in shipborne HFSWR, especially for detecting extended targets. [ABSTRACT FROM AUTHOR]

Spectrum sensing (SS) technology is essential for cognitive radio (CR) networks to effectively identify and utilize idle spectrum resources. Due to the influence of noise characteristics in the channel, providing accurate sensing results is challenging. In order to improve the performance of SS under non-Gaussian noise and overcome the limitations of existing methods that are mostly based on a single feature, we propose a novel time-frequency cross fusion network (TFCFN). Specifically, we utilize gated recurrent units (GRU) to capture long-term dependencies in the time domain on the original signals, meanwhile, we perform a fast Fourier transform (FFT) on the original signals to obtain the frequency domain information, and subsequently use convolutional neural networks (CNN) to extract the local spatial features in the frequency domain. Ultimately, these time-domain and frequency-domain features are dynamically fused through a cross-attention mechanism to construct more comprehensive and robust features for signal classification. We use generalized Gaussian distribution (GGD) as the noise model and reconstruct the RadioML2016.10a dataset to explore the performance under various noise conditions. The experimental results show that compared with the baseline methods, TFCFN exhibits better detection ability and maintains lower complexity in both Gaussian and non-Gaussian noise environments. Notably, when the shape parameter of GGD is set to 0.5 and the signal-to-noise ratio (SNR) of the received signal is -16dB, it can maintain the probability of false alarm ( P f ) of 10% while still ensuring the probability of detection ( P d ) of over 90%. [ABSTRACT FROM AUTHOR]

Heart sound auscultation plays a crucial role in the early diagnosis of cardiovascular diseases. In recent years, great achievements have been made in the automatic classification of heart sounds, but most methods are based on segmentation features and traditional classifiers and do not fully exploit existing deep networks. This paper proposes a cardiac audio classification method based on image expression of multidimensional features (CACIEMDF). First, a 102-dimensional feature vector is designed by combining the characteristics of heart sound data in the time domain, frequency domain and statistical domain. Based on the feature vector, a two-dimensional feature projection space is constructed by PCA dimensionality reduction and the convex hull algorithm, and 102 pairs of coordinate representations of the feature vector in the two-dimensional space are calculated. Each one-dimensional component of the feature vector corresponds to a pair of 2D coordinate representations. Finally, the one-dimensional feature component value and its divergence into categories are used to fill the three channels of a color image, and a Gaussian model is used to dye the image to enrich its content. The color image is sent to a deep network such as ResNet50 for classification. In this paper, three public heart sound datasets are fused, and experiments are conducted using the above methods. The results show that for the two-classification/five-classification task of heart sounds, the method in this paper can achieve a classification accuracy of 95.68%/94.53% when combined with the current deep network. [ABSTRACT FROM AUTHOR]

Background: Epilepsy is a prevalent neurological disorder in which seizures cause recurrent episodes of unconsciousness or muscle convulsions, seriously affecting the patient's work, quality of life, and health and safety. Timely prediction of seizures is critical for patients to take appropriate therapeutic measures. Accurate prediction of seizures remains a challenge due to the complex and variable nature of EEG signals. The study proposes an epileptic seizure model based on a multidimensional Transformer with recurrent neural network(LSTM-GRU) fusion for seizure classification of EEG signals. Methodology: Firstly, a short-time Fourier transform was employed in the extraction of time-frequency features from EEG signals. Second, the extracted time-frequency features are learned using the Multidimensional Transformer model. Then, LSTM and GRU are then used for further learning of the time and frequency characteristics of the EEG signals. Next, the output features of LSTM and GRU are spliced and categorized using the gating mechanism. Subsequently, seizure prediction is conducted. Results: The model was tested on two datasets: the Bonn EEG dataset and the CHB-MIT dataset. On the CHB-MIT dataset, the average sensitivity and average specificity of the model were 98.24% and 97.27%, respectively. On the Bonn dataset, the model obtained about 99% and about 98% accuracy on the binary classification task and the tertiary upper classification task, respectively. Conclusion: The findings of the experimental investigation demonstrate that our model is capable of exploiting the temporal and frequency characteristics present within EEG signals. [ABSTRACT FROM AUTHOR]

Sleep apnea (SA) is a prevalent disorder characterized by recurrent events of nocturnal apnea. Polysomnography (PSG) represents the gold standard for SA diagnosis. This laboratory-based procedure is complex and costly, and less cumbersome wearable devices have been proposed for SA detection and monitoring. A novel textile multi-sensor monitoring belt recording electrocardiogram (ECG) and breathing frequency (BF) measured by thorax excursion was developed and tested in a sleep laboratory for validation purposes. The aim of the current study was to evaluate the diagnostic performance of ECG-derived heart rate variability and BF-derived breathing rate variability and their combination for the detection of sleep apnea in a population of patients with a suspicion of SA. Fifty-one patients with a suspicion of SA were recruited in the sleep laboratory of the Cantonal Hospital St. Gallen. Patients were equipped with the monitoring belt and underwent a single overnight laboratory-based PSG. In addition, some patients further tested the monitoring belt at home. The ECG and BF signals from the belt were compared to PSG signals using the Bland-Altman methodology. Heart rate and breathing rate variability analyses were performed. Features derived from these analyses were used to build a support vector machine (SVM) classifier for the prediction of SA severity. Model performance was assessed using receiver operating characteristics (ROC) curves. Patients included 35 males and 16 females with a median age of 49 years (range: 21 to 65) and a median apnea-hypopnea index (AHI) of 33 (IQR: 16 to 58). Belt-derived data provided ECG and BF signals with a low bias and in good agreement with PSG-derived signals. The combined ECG and BF signals improved the classification accuracy for SA (area under the ROC curve: 0.98; sensitivity and specificity greater than 90%) compared to single parameter classification based on either ECG or BF alone. This novel wearable device combining ECG and BF provided accurate signals in good agreement with the gold standard PSG. Due to its unobtrusive nature, it is potentially interesting for multi-night assessments and home-based patient follow-up. [ABSTRACT FROM AUTHOR]

Large-scale datasets inevitably contain noisy labels, which induces weak performance of deep neural networks (DNNs). Many existing methods focus on loss and regularization tricks, as well as characterizing and modelling differences between noisy and clean samples. However, taking advantage of information from different extents of distortion in latent feature space, is less explored and remains challenging. To solve this problem, we analyze characteristic distortion extents of different high-dimensional features, achieving the conclusion that features vary in their degree of deformation in their correlations with respect to categorical variables. Aforementioned disturbances on features not only reduce sensitivity and contribution of latent features to classification, but also bring obstacles into generating decision boundaries. To mitigate these issues, we propose class sensitivity feature extractor (CSFE) and T-type generative classifier (TGC). Based on the weighted Mahalanobis distance between conditional and unconditional cumulative distribution function after variance-stabilizing transformation, CSFE realizes high quality feature extraction through evaluating class-wise discrimination ability and sensitivity to classification. TGC introduces student-t estimator to clustering analysis in latent space, which is more robust in generating decision boundaries while maintaining equivalent efficiency. To alleviate the cost of retraining a whole DNN, we propose an ensemble model to simultaneously generate robust decision boundaries and train the DNN with the improved CSFE named SoftCSFE. Extensive experiments on three datasets, which are the RML2016.10a dataset, UCR Time Series Classification Archive dataset and a real-world dataset Clothing1M, show advantages of our methods. [ABSTRACT FROM AUTHOR]

Automatic modulation recognition (AMR) methods used in advanced wireless communications systems can identify unknown signals without requiring reference information. However, the acceptance of these methods depends on the accuracy, number of parameters, and computational complexity. This study proposes a hybrid convolutional transformer classifier (HCTC) for the classification of unknown signals. The proposed method utilizes a three-stage framework to extract features from in-phase/quadrature (I/Q) signals. In the first stage, spatial features are extracted using a convolutional layer. In the second stage, temporal features are extracted using a transformer encoder. In the final stage, the features are mapped using a deep-learning network. The proposed HCTC method is investigated using the benchmark RadioML database and compared with state-of-the-art methods. The experimental results demonstrate that the proposed method achieves a better performance in modulation signal classification. Additionally, the performance of the proposed method is evaluated when applied to different batch sizes and model configurations. Finally, open issues in modulation recognition research are addressed, and future research perspectives are discussed. [ABSTRACT FROM AUTHOR]

The accuracy of Direction of Arrival (DOA) estimation primarily depends on the precision of the data. When the receiver uses a low-precision analog-to-digital converter (ADC), traditional DOA estimation algorithms exhibit poor accuracy. To face the challenge of multi-target DOA estimation in scenarios with low-precision ADC quantized sampling, this paper proposes a novel DOA estimation algorithm for quantized signals based on classification problems. A deep learning network was constructed using Deep Neural Networks (DNNs) and Convolutional Neural Networks (CNNs), divided into the quantized signal recovery framework and the DOA estimation framework. The DNN network is utilized to recover signals that have undergone low-precision quantization, while the CNN network addresses the classification problem to estimate the DOA from received data with an unknown number of signal sources. A comprehensive analysis of the impact of signal-to-noise ratio (SNR), the number of array elements, and the number of quantization bits on the proposed algorithm was conducted. Simulation results indicate that the proposed algorithm exhibits superior DOA estimation performance in low-precision scenarios, characterized by reduced computational complexity, thereby facilitating real-time DOA estimation. [ABSTRACT FROM AUTHOR]

Automation and digitalization in various industries towards the Industry 4.0/5.0 paradigms are rapidly progressing thanks to the use of sensors, Industrial Internet of Things (IIoT), and advanced fifth generation (5G) and sixth generation (6G) mobile networks supported by simulation and automation of processes using artificial intelligence (AI) and machine learning (ML). Ensuring the continuity of operations under different conditions is becoming a key factor. One of the most frequently requested solutions is currently predictive maintenance, i.e., the simulation and automation of maintenance processes based on ML. This article aims to extract the main trends in the area of ML-based predictive maintenance present in studies and publications, critically evaluate and compare them, and define priorities for their research and development based on our own experience and a literature review. We provide examples of how BCI-controlled predictive maintenance due to brain–computer interfaces (BCIs) play a transformative role in AI-based predictive maintenance, enabling direct human interaction with complex systems. [ABSTRACT FROM AUTHOR]

Transformer models have revolutionized fields like Natural Language Processing (NLP) by enabling machines to accurately understand and generate human language. However, these models' inherent complexity and limited interpretability pose barriers to their broader adoption. To address these challenges, we propose a simplified query-only attention mechanism specifically for encoder-based transformer models to reduce complexity and improve interpretability. Unlike conventional attention mechanisms, which rely on query (Q), key (K), and value (V) vectors, our method uses only the Q vector for attention calculation. This approach reduces computational complexity while maintaining the model's ability to capture essential relationships, enhancing interpretability. We evaluated the proposed query-only attention on an EEG conformer model, a state-of-the-art architecture for EEG signal classification. We demonstrated that it performs comparably to the original QKV attention mechanism, while simplifying the model's architecture. Our findings suggest that query-only attention offers a promising direction for the development of more efficient and interpretable transformer-based models, with potential applications across various domains beyond NLP. [ABSTRACT FROM AUTHOR]

This study introduces a novel deep learning-based approach for classifying poultry audio signals, incorporating a custom Burn Layer to enhance model robustness. The methodology integrates digital audio signal processing, convolutional neural networks (CNNs), and the innovative Burn Layer, which injects controlled random noise during training to reinforce the model's resilience to input signal variations. The proposed architecture is streamlined, with convolutional blocks, densely connected layers, dropout, and an additional Burn Layer to fortify robustness. The model demonstrates efficiency by reducing trainable parameters to 191,235, compared to traditional architectures with over 1.7 million parameters. The proposed model utilizes a Burn Layer with burn intensity as a parameter and an Adamax optimizer to optimize and address the overfitting problem. Thorough evaluation using six standard classification metrics showcases the model's superior performance, achieving exceptional sensitivity (96.77%), specificity (100.00%), precision (100.00%), negative predictive value (NPV) (95.00%), accuracy (98.55%), F1 score (98.36%), and Matthew's correlation coefficient (MCC) (95.88%). This research contributes valuable insights into the fields of audio signal processing, animal health monitoring, and robust deep-learning classification systems. The proposed model presents a systematic approach for developing and evaluating a deep learning-based poultry audio classification system. It processes raw audio data and labels to generate digital representations, utilizes a Burn Layer for training variability, and constructs a CNN model with convolutional blocks, pooling, and dense layers. The model is optimized using the Adamax algorithm and trained with data augmentation and early-stopping techniques. Rigorous assessment on a test dataset using standard metrics demonstrates the model's robustness and efficiency, with the potential to significantly advance animal health monitoring and disease detection through audio signal analysis. [ABSTRACT FROM AUTHOR]

The modulation classification technology for radar intra-pulse signals is important in the electronic countermeasures field. As the high quality labeled radar signals are difficult to be captured in the real applications, the signal modulation classification base on the limited number of labeled samples is playing a more and more important role. To relieve the requirement of the labeled samples, many self-supervised learning (SeSL) models exist. However, as they cannot fully explore the information of the labeled samples and rely significantly on the unlabeled samples, highly time-consuming processing of the pseudo-labels of the unlabeled samples is caused. To solve these problems, a supervised learning (SL) model, using the contrastive learning (CL) method (SL-CL), is proposed in this paper, which achieves a high classification accuracy, even adopting limited number of labeled training samples. The SL-CL model uses a two-stage training structure, in which the CL method is used in the first stage to effectively capture the features of samples, then the multilayer perceptron is applied in the second stage for the classification. Especially, the supervised contrastive loss is constructed to fully exploring the label information, which efficiently increases the classification accuracy. In the experiments, the SL-CL outperforms the comparison models in the situation of limited number of labeled samples available, which reaches 94% classification accuracy using 50 samples per class at 5 dB SNR. [ABSTRACT FROM AUTHOR]

This article presents a systematic review using PRISMA methodology to explore trends in the use of machine and deep learning in diagnosing and detecting Alzheimer's disease using electroencephalography. This review covers studies published between 2013 and 2023, drawing on three leading academic databases: Scopus, Web of Science, and PubMed. The validity of the databases is evaluated considering essential factors such as the arrangement of EEG electrodes, data acquisition methodologies, and the number of participants. Additionally, the specific properties of the databases used in the research are highlighted, including EEG signal classification, filtering, segmentation approaches, and selected features. Finally, the performance metrics of the classification algorithms are evaluated, especially the accuracy achieved, offering a comprehensive view of the current state and future trends in the use of these technologies for the diagnosis of Alzheimer's disease. [ABSTRACT FROM AUTHOR]

In this paper, a method for the classification of anomalous heartbeats from compressed ECG signals is proposed. The method operating on signals acquired by compressed sensing is based on a feature extraction stage consisting of the evaluation of the Discrete Cosine Transform (DCT) coefficients of the compressed signal and a classification stage performed by means of a set of k-nearest neighbor ensemble classifiers. The method was preliminarily tested on five classes of anomalous heartbeats, and it achieved a classification accuracy of 99.40%. [ABSTRACT FROM AUTHOR]

Electroretinography (ERG) is a non-invasive method of assessing retinal function by recording the retina's response to a brief flash of light. This study focused on optimizing the ERG waveform signal classification by utilizing Short-Time Fourier Transform (STFT) spectrogram preprocessing with a machine learning (ML) decision system. Several window functions of different sizes and window overlaps were compared to enhance feature extraction concerning specific ML algorithms. The obtained spectrograms were employed to train deep learning models alongside manual feature extraction for more classical ML models. Our findings demonstrated the superiority of utilizing the Visual Transformer architecture with a Hamming window function, showcasing its advantage in ERG signal classification. Also, as a result, we recommend the RF algorithm for scenarios necessitating manual feature extraction, particularly with the Boxcar (rectangular) or Bartlett window functions. By elucidating the optimal methodologies for feature extraction and classification, this study contributes to advancing the diagnostic capabilities of ERG analysis in clinical settings. [ABSTRACT FROM AUTHOR]

The rapid evolution of drone technology has introduced unprecedented challenges in security, particularly concerning the threat of unconventional drone and swarm attacks. In order to deal with threats, drones need to be classified by intercepting their Radio Frequency (RF) signals. With the arrival of Sixth Generation (6G) networks, it is required to develop sophisticated methods to properly categorize drone signals in order to achieve optimal resource sharing, high-security levels, and mobility management. However, deep ensemble learning has not been investigated properly in the case of 6G. It is anticipated that it will incorporate drone-based BTS and cellular networks that, in one way or another, may be subjected to jamming, intentional interferences, or other dangers from unauthorized UAVs. Thus, this study is conducted based on Radio Frequency Fingerprinting (RFF) of drones identified to detect unauthorized ones so that proper actions can be taken to protect the network's security and integrity. This paper proposes a novel method—a Composite Ensemble Learning (CEL)-based neural network—for drone signal classification. The proposed method integrates wavelet-based denoising and combines automatic and manual feature extraction techniques to foster feature diversity, robustness, and performance enhancement. Through extensive experiments conducted on open-source benchmark datasets of drones, our approach demonstrates superior classification accuracies compared to recent benchmark deep learning techniques across various Signal-to-Noise Ratios (SNRs). This novel approach holds promise for enhancing communication efficiency, security, and safety in 6G networks amidst the proliferation of drone-based applications. [ABSTRACT FROM AUTHOR]

Copyright of Informatica (03505596) is the property of Slovene Society Informatika and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The ability to effectively classify human emotion states is critically important for human‐computer or human‐robot interactions. However, emotion classification with physiological signals is still a challenging problem due to the diversity of emotion expression and the characteristic differences in different modal signals. A novel learning‐based network architecture is presented that can exploit four‐modal physiological signals, electrocardiogram, electrodermal activity, electromyography, and blood volume pulse, and make a classification of emotion states. It features two kinds of attention modules, feature‐level, and semantic‐level, which drive the network to focus on the information‐rich features by mimicking the human attention mechanism. The feature‐level attention module encodes the rich information of each physiological signal. While the semantic‐level attention module captures the semantic dependencies among modals. The performance of the designed network is evaluated with the open‐source Wearable Stress and Affect Detection dataset. The developed emotion classification system achieves an accuracy of 83.88%. Results demonstrated that the proposed network could effectively process four‐modal physiological signals and achieve high accuracy of emotion classification. [ABSTRACT FROM AUTHOR]

Drainage tunnel stability is crucial for engineering project safety (e.g., mine engineering and dams), and rockfall events and water release are key indicators of drainage tunnel stability. To address this, we developed a monitoring system to simulate drainage tunnel intrusions based on distributed acoustic sensing (DAS), and we obtained typical characteristics of events like rockfall events and water release. Given the multitude of DAS signal feature parameters and challenges, such as high-dimensional features impacting the classification accuracy of machine learning, we proposed an identification method for drainage tunnel intrusion events using principal component analysis (PCA) and neural networks. PCA reveals that amplitude-related parameters—amplitude, mean amplitude, and energy—significantly contribute to DAS signal classification, reducing the feature parameter dimensions by 54.8%. The accuracy of intrusion event classification improves with PCA-processed data compared to unprocessed data, with overall accuracy rates of 79.1% for rockfall events and 72.7% for water release events. Additionally, the artificial neural network model outperforms the Bayesian and logistic regression models, demonstrating that ANN has advantages in handling complex models for intrusion event classification. [ABSTRACT FROM AUTHOR]

In this research, five systems were developed to classify four distinct motor functions—forward hand movement (FW), grasp (GP), release (RL), and reverse hand movement (RV)—from EEG signals, using the WAY-EEG-GAL dataset where participants performed a sequence of hand movements. During preprocessing, band-pass filtering was applied to remove artifacts and focus on the mu and beta frequency bands. The initial system, a preliminary study model, explored the overall framework of EEG signal processing and classification, utilizing time-domain features such as variance and frequency-domain features such as alpha and beta power, with a KNN model for classification. Insights from this study informed the development of a baseline system, which innovatively combined the common spatial patterns (CSP) method with continuous wavelet transform (CWT) for feature extraction and employed a GoogLeNet classifier with transfer learning. This system classified six unique pairs of events derived from the four motor functions, achieving remarkable accuracy, with the highest being 99.73% for the GP–RV pair and the lowest 80.87% for the FW–GP pair in intersubject classification. Building on this success, three additional systems were developed for four-way classification. The final model, ML-CSP-OVR, demonstrated the highest intersubject classification accuracy of 78.08% using all combined data and 76.39% for leave-one-out intersubject classification. This proposed model, featuring a novel combination of CSP-OVR, CWT, and GoogLeNet, represents a significant advancement in the field, showcasing strong potential as a general system for motor imagery (MI) tasks that is not dependent on the subject. This work highlights the prominence of the research contribution by demonstrating the effectiveness and robustness of the proposed approach in achieving high classification accuracy across different motor functions and subjects. [ABSTRACT FROM AUTHOR]

Epilepsy is a disorder that interferes with regular brain activity and can occasionally cause seizures, odd sensations, and momentary unconsciousness. Epilepsy is frequently diagnosed using electroencephalograph (EEG) records, although conventional analysis is subjective and prone to error. The dynamic and non‐stationary nature of EEG structure restricted the performance of Deep Learning (DL) approaches used in earlier work to improve EEG classification. Our multi‐channel EEG classification model, dubbed LConvNet in this paper, combines Convolutional Neural Networks (CNN) for extracting spatial features and Long Short‐Term Memory (LSTM) for identifying temporal dependencies. To discriminate between epileptic and healthy EEG signals, the model is trained using open‐source secondary EEG data from Temple University Hospital (TUH). Our model outperformed other EEG classification models employed in comparable tasks, such as EEGNet, DeepConvNet, and ShallowConvNet, which had accuracy rates of 86%, 96%, and 78%, respectively. Our model attained an amazing accuracy rate of 97%. During additional testing, our model also displayed excellent performance in trainability, scalability, and parameter efficiency. [ABSTRACT FROM AUTHOR]

Cardiac arrhythmia identification and categorization are crucial for prompt treatment and better patient outcomes. Arrhythmia identification is the main focus of this study's temporal attention network (TAN)-based multiclass categorization of varied-length electrocardiogram (ECG) data. The suggested TAN is designed to handle variable-duration ECG signals, making it ideal for real-time monitoring. The TAN uses a dynamic snippet extraction approach to choose meaningful ECG segments to ensure the model captures essential properties despite the constraints of processing such heterogeneous data. Training and assessment use a large dataset of atrial fibrillation, ventricular, and supraventricular arrhythmias. The TAN outperforms current approaches in multiclass early arrhythmia classification and is very accurate. Concatenating EfficientNet with CNN layer helped overcome different data and variable-length signals. High accuracy: 98% of normal, 97.1% of atrial fibrillation (AF), 98% of other, and 98% of noisy using the proposed CEEC model. Early arrhythmia diagnosis has improved due to the TAN's ability to effectively identify varied-length ECG data and give interpretability. It enables quicker interventions, personalised treatment plans, and improved arrhythmia control, which can greatly benefit patient care. [ABSTRACT FROM AUTHOR]

Copyright of Przegląd Elektrotechniczny is the property of Przeglad Elektrotechniczny and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Quantum reservoir computing (QRC) has been proposed as a paradigm for performing machine learning with quantum processors where the training takes place in the classical domain, avoiding the issue of barren plateaus in parameterized-circuit quantum neural networks. It is natural to consider using a quantum processor based on microwave superconducting circuits to classify microwave signals that are analog—continuous in time. However, while there have been theoretical proposals of analog QRC, to date QRC has been implemented using the circuit model—imposing a discretization of the incoming signal in time. In this paper we show how a quantum superconducting circuit comprising an oscillator coupled to a qubit can be used as an analog quantum reservoir for a variety of classification tasks, achieving high accuracy on all of them. Our work demonstrates processing of ultra-low-power microwave signals within our superconducting circuit, a step towards achieving a quantum sensing-computational advantage on impinging microwave signals. Quantum reservoir computing might in principle give advantages in solving signal classification tasks, but current implementations usually incur in the digitalization bottleneck. Here, the authors demonstrate an implementation dealing directly with the analogue MW signals. [ABSTRACT FROM AUTHOR]

The classification of electroencephalogram (EEG) signals derived from motor imagery (MI) has always been a hot topic in the field of brain–computer interfaces. Due to its ability to handle the nonstationary and uncertain information contained in EEG signals, the Takagi–Sugeno–Kang fuzzy system (TSK-FS) has become an advantageous classification algorithm. To train a fuzzy system with strong discrimination capabilities from EEG data interspersed with redundant information, this paper proposes a TSK-FS modeling method based on low-rank sparse subspace learning (TSK-LSSL). This method focuses on consequent parameter learning, which transforms the traditional consequent parameter learning strategy into low-rank subspace and sparse subspace learning processes. Low-rank subspace learning is used to mine the global structural information of data and effectively reduce the number of fuzzy rules. During sparse subspace learning, ℓ2,1-norm regularization is used to constrain the consequent parameters and causes the number of redundant consequent parameters to be zero, thereby simplifying the fuzzy rules. In addition, a local boundary term based on graph matrices is embedded into the objective function to mine the local structural information of the given data. TSK-LSSL simplifies the number of rules and the consequent part of the fuzzy rules. It exhibits good classification performance on two BCI Competition databases. [ABSTRACT FROM AUTHOR]

The alarm calls of vervet monkeys (Chlorocebus pygerythrus) have been the subject of considerable focus by researchers, owing primarily to the purported referential qualities of different alarm call types. With this focus on reference, acoustic variation among calls elicited by the same range of predators has typically been overlooked. Specifically, at least one type of alarm call—the terrestrial alarm—was described over 50 years ago as being acoustically distinct between males and females—a description that has largely eluded more systematic scrutiny. Here, we provide a quantitative acoustic analysis and comparison of terrestrial alarm calls produced by adult male and female vervet monkeys. We use a random forest model to determine which acoustic variables best distinguish between the calls of males and females, and use an unsupervised clustering technique to objectively determine whether alarms produced by each sex fall into discrete types. We found that the calls of males and females differed most in frequency‐based parameters, with male alarms containing more energy at lower frequencies relative to females. Calls produced by males were also of longer duration, and consisted of longer individual call elements relative to female calls. While calls generally fell into clusters associated with either male or female alarms, we found that some fell into atypical clusters given the caller's sex, and that the clusters themselves showed evidence of intergradation. We discuss these results in terms of potential differences in the function of, and motivation for, calling by males and females. We emphasize the need for a more holistic approach to the classification of vocal signals that considers contextual, functional, and structural variation. [ABSTRACT FROM AUTHOR]

Signal detection and classification tasks, especially in the realm of audio, suffer from difficulties in capturing long short-term dependencies and effectively utilizing samples. Firstly, audio signal detection and classification need to classify audio signals and detect their onset and offset times; therefore, obtaining long short-term dependencies is necessary. The methods based on RNNs have high time complexity and dilated convolution-based methods suffer from the "gridding issue" challenge; thus, the HDC-Inception module is proposed to efficiently extract long short-term dependencies. Combining the advantages of the Inception module and a hybrid dilated convolution (HDC) framework, the HDC-Inception module can both alleviate the "gridding issue" and obtain long short-term dependencies. Secondly, datasets have large numbers of silent segments and too many samples for some signal types, which are redundant and less difficult to detect, and, therefore, should not be overly prioritized. Thus, selecting effective samples and guiding the training based on them is of great importance. Inspired by support vector machine (SVM), combining soft margin SVM and cross-entropy loss (CE loss), the soft margin CE loss is proposed. Soft margin CE loss can adaptively select support vectors (effective samples) in datasets and guide training based on the selected samples. To utilize datasets more sufficiently, a hybrid CE loss is proposed. Using the benefits of soft margin CE loss and CE loss, hybrid CE loss guides the training with all samples and gives weight to support vectors. Soft margin CE loss and hybrid CE loss can be extended to most classification tasks and offer a wide range of applications and great potential. [ABSTRACT FROM AUTHOR]

Non-line-of-sight (NLOS) errors significantly impact the accuracy of ultra-wideband (UWB) indoor positioning, posing a major barrier to its advancement. This study addresses the challenge of effectively distinguishing line-of-sight (LOS) from NLOS signals to enhance UWB positioning accuracy. Unlike existing research that focuses on optimizing deep learning network structures, our approach emphasizes the optimization of model parameters. We introduce a chaotic map for the initialization of the population and integrate a subtraction-average-based optimizer with a dynamic exploration probability to enhance the Snake Search Algorithm (SSA). This improved SSA optimizes the initial weights and thresholds of backpropagation (BP) neural networks for signal classification. Comparative evaluations with BP, Particle Swarm Optimizer–BP (PSO-BP), and Snake Optimizer–PB (SO-BP) models—performed using three performance metrics—demonstrate that our LTSSO-BP model achieves superior stability and accuracy, with classification accuracy, recall, and F1 score values of 90%, 91.41%, and 90.25%, respectively. [ABSTRACT FROM AUTHOR]

A deep-seated landslide could release numerous microseismic signals from creep-slip movement, which includes a rock-soil slip from the slope surface and a rock-soil shear rupture in the subsurface. Machine learning can effectively enhance the classification of microseismic signals in landslide seismic monitoring and interpret the mechanical processes of landslide motion. In this paper, eight sets of triaxial seismic sensors were deployed inside the deep-seated landslide, Jiuxianping, China, and a large number of microseismic signals related to the slope movement were obtained through 1-year-long continuous monitoring. All the data were passed through the seismic event identification mode, the ratio of the long-time average and short-time average. We selected 11 days of data, manually classified 4131 data into eight categories, and created a microseismic event database. Classical machine learning algorithms and ensemble learning algorithms were tested in this paper. In order to evaluate the seismic event classification performance of each algorithmic model, we evaluated the proposed algorithms through the dimensions of the accuracy, precision, and recall of each model. The validation results demonstrated that the best performing decision tree algorithm among the classical machine learning algorithms had an accuracy of 88.75%, while the ensemble algorithms, including random forest, Gradient Boosting Trees, Extreme Gradient Boosting, and Light Gradient Boosting Machine, had an accuracy range from 93.5% to 94.2% and also achieved better results in the combined evaluation of the precision, recall, and F1 score. The specific classification tests for each microseismic event category showed the same results. The results suggested that the ensemble learning algorithms show better results compared to the classical machine learning algorithms. [ABSTRACT FROM AUTHOR]

The detection performance of radar is significantly impaired by active jamming and mutual interference from other radars. This paper proposes a radio signal modulation recognition method to accurately recognize these signals, which helps in the jamming cancellation decisions. Based on the ensemble learning stacking algorithm improved by meta-feature enhancement, the proposed method adopts random forests, K-nearest neighbors, and Gaussian naive Bayes as the base-learners, with logistic regression serving as the meta-learner. It takes the multi-domain features of signals as input, which include time-domain features including fuzzy entropy, slope entropy, and Hjorth parameters; frequency-domain features, including spectral entropy; and fractal-domain features, including fractal dimension. The simulation experiment, including seven common signal types of radar and active jamming, was performed for the effectiveness validation and performance evaluation. Results proved the proposed method's performance superiority to other classification methods, as well as its ability to meet the requirements of low signal-to-noise ratio and few-shot learning. [ABSTRACT FROM AUTHOR]

A method for recognizing infrasound acoustic signals for two types of sources based on the analysis of the shape of their wavelet spectra is proposed. The idea of constructing this form is based on the principal component method. Morphological image analysis methods are used to search for characteristic areas. The proposed method makes it possible to effectively solve the problem of multiclass classification of acoustic signals. [ABSTRACT FROM AUTHOR]

Over the past decade, there have been great advancements in radio frequency sensor technology for human–computer interaction applications, such as gesture recognition, and human activity recognition more broadly. While there is a significant amount of study on these topics, in most cases, experimental data are acquired in controlled settings by directing participants what motion to articulate. However, especially for communicative motions, such as sign language, such directed data sets do not accurately capture natural, in situ articulations. This results in a difference in the distribution of directed American Sign Language (ASL) versus natural ASL, which severely degrades natural sign language recognition in real‐world scenarios. To overcome these challenges and acquire more representative data for training deep models, the authors develop an interactive gaming environment, ChessSIGN, which records video and radar data of participants as they play the game without any external direction. The authors investigate various ways of generating synthetic samples from directed ASL data, but show that ultimately such data does not offer much improvement over just initialising using imagery from ImageNet. In contrast, an interactive learning paradigm is proposed by the authors in which model training is shown to improve as more and more natural ASL samples are acquired and augmented via synthetic samples generated from a physics‐aware generative adversarial network. The authors show that the proposed approach enables the recognition of natural ASL in a real‐world setting, achieving an accuracy of 69% for 29 ASL signs—a 60% improvement over conventional training with directed ASL data. [ABSTRACT FROM AUTHOR]

Radar High‐Resolution Range Profile (HRRP) has great potential for target recognition because it can provide target structural information. Existing work commonly applies deep learning to extract deep features from HRRPs and achieve impressive recognition performance. However, most approaches are unable to distinguish between the target and non‐target regions in the feature extraction process and do not fully consider the impact of background noise, which is harmful to recognition, especially at low signal‐to‐noise ratios (SNR). To tackle these problems, the authors propose a radar HRRP target recognition framework termed Adaptive Soft Threshold Transformer (ASTT), which is composed of a patch embedding (PE) layer, ASTT blocks, and Discrete Wavelet Patch Merging (DWPM) layers. Given the limited semantic information of individual range cells, the PE layer integrates nearby isolated range cells into semantically explicit target structure patches. Thanks to its convolutional layer and attention mechanism, the ASTT blocks assign a weight to each patch to locate the target areas in the HRRP while capturing local features and constructing sequence correlations. Moreover, the ASTT block efficiently filters noise features in combination with a soft threshold function to further enhance the recognition performance at low SNR, where the threshold is adaptively determined. Utilising the reversibility of the discrete wavelet transform, the DWPM layer efficiently eliminates the loss of valuable information during the pooling process. Experiments based on simulated and measured datasets show that the proposed method has excellent target recognition performance, noise robustness, and small‐scale range shift robustness. [ABSTRACT FROM AUTHOR]

To overcome AMR s limited generalization and low classification accuracy in non-cooperative communication contexts with low signal-to-noise ratio, this paper proposed a model named CGDNN, which integrated CNN, GRU and deep neural networks. To mitigate noise impact on modulation detection, this paper initially denoised I/Q sampled signal using wavelet thresholding. Subsequently, this paper utilized CNN and GRU for extracting spatial and temporal features from signals before proceeding to classification through fully connected layers. Besides enhancing AMR performance, the CGDNN model significantly reduced computational complexity compared to competitors. Experiment results demonstrate an average recognition accuracy of 64.32% on the RML2016. 10b dataset, with an enhancement in signal classification accuracy from 12 dB to 0 dB. Moreover, the model substantially decreased confusion between 16QAM and 64QAM, achieving a peak recognition accuracy of 93.9% at 18 dB. CGDNN model effectively improved AMR detection accuracy under low signal-to-noise ratio conditions and enhanced model training efficiency. [ABSTRACT FROM AUTHOR]

In recent years, emotion recognition based on EEG signals has received significant attention and research interest. EEG signals have the advantages of universality, spontaneity, and difficulty in deception, making them capable of accurately reflecting genuine emotional states. In this field, researchers have conducted binary (high/low) and ternary (low/medium/high) classification studies on the valence and arousal levels in the DEAP dataset. However, in order to better identify deep and intrinsic emotions, a clear definition of emotions becomes particularly important. Therefore, this study refers to Russell's Circumplex Model, which arranges emotions in a circular manner based on their valence and arousal levels. The study proposes placing emotion labels from the DEAP dataset within the two-dimensional emotional space of the circumplex model. Emotions are defined as four labels - Excited, Afraid, Sad, and Relaxed - based on a linear distribution of valence and arousal levels. Furthermore, a hybrid deep learning model combining CNN and Transformer is proposed for multi-scale spatial-temporal feature extraction. This model is employed for the classification of the four emotions. Finally, the model achieves an average accuracy of 91.26% on the four-class emotion classification task. [ABSTRACT FROM AUTHOR]

Electroencephalography (EEG), used to record the random electrical activity in brain, is a known medical test. In this test, a graphical waveform is obtained by measuring the electrical activity of the cells. In the medical world, the relationship between epilepsy and EEG can be understood by examining changes in brain activity during or between epileptic seizures. EEG is a useful tool in the early treatment and diagnosis of epilepsy. Whether seizures, generally known as abnormal electrical discharges in brain cells, are of epileptic origin, comes to light through EEG. The main goal of our study was to demonstrate the EEG rhythm effectiveness for the diagnosis of epilepsy in EEG data obtained from the epilepsy center of Bonn Freiburg University Hospital. Time domain feature extraction of EEG band classification results was examined in detail against the classification results of frequency domain feature extraction of EEG rhythms in healthy subjects and subjects with epilepsy. By extracting effective features from EEG data in both time and frequency domains, the k nearest neighbor (KNN) algorithm was used for the time and frequency domain. It cannot be overlooked that among the four methods used for performance evaluation in the designed model, the classification success of frequency domain features was more successful than that of time domain features. Using the KNN algorithm, healthy individuals and epilepsy patients with seizures were classified with 100% success. [ABSTRACT FROM AUTHOR]

Myocardial infarction (MI) poses a significant clinical challenge, necessitating expeditious and precise detection to mitigate potentially fatal outcomes. Current MI diagnosis predominantly relies on electrocardiography (ECG); however, it is fraught with limitations, including inter-observer variability and a reliance on expert interpretation. This study introduces an automated MI detection framework that capitalizes on hybrid signal processing methodologies and deterministic learning theory. The initial step involves the extraction of the Shannon energy envelope (SEE) and its derivative from a single-lead ECG. Integration of the SEE into the ECG's phase portrait provides a means to capture the underlying nonlinear system dynamics. Subsequently, the application of fast and adaptive multivariate empirical mode decomposition (FA-MVEMD) yields discriminative features originating from the most energetically dominant intrinsic mode components (IMFs) within the SEE. Profound dissimilarities are discernible between ECG signals recorded from healthy subjects and those afflicted with MI. In the subsequent phase, deterministic learning theory, implemented through neural networks, is employed to facilitate the classification of ECG signals into two distinct groups. The method's efficacy is meticulously evaluated using the PTB diagnostic ECG database, resulting in a noteworthy average classification accuracy of 99.21 % within a tenfold cross-validation framework. In summation, the findings affirm that the proposed features not only complement conventional ECG attributes but also align closely with the underlying dynamics of the ECG system, ultimately fortifying the automatic detection of MI. The imperative requirement for early and accurate MI diagnosis is addressed through our approach, offering a robust and dependable means to fulfill this pivotal clinical need. [ABSTRACT FROM AUTHOR]

Heart sounds and murmur provide crucial diagnosis information for valvular heart diseases (VHD). A phonocardiogram (PCG) combined with modern digital processing techniques provides a complementary tool for clinicians. This article proposes a benchmark different time–frequency representations, which are spectograms, mel-spectograms and cochleagrams for obtaining images, in addition to the use of two interpolation techniques to improve the quality of the images, which are bicubic and Lanczos. Deep features are extracted from a pretrained model called VGG16, and for feature reduction, the Boruta algorithm is applied. To evaluate the models and obtain more precise results, nested cross-validation is used. The best results achieved in this study were for the cochleagram with 99.2% accuracy and mel-spectogram representation with the bicubic interpolation technique, which reached 99.4% accuracy, both having a support vector machine (SVM) as a classifier algorithm. Overall, this study highlights the potential of time–frequency representations of PCG signals combined with modern digital processing techniques and machine learning algorithms for accurate diagnosis of VHD. [ABSTRACT FROM AUTHOR]

Copyright of Przegląd Elektrotechniczny is the property of Przeglad Elektrotechniczny and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract Over the past decade, there have been great advancements in radio frequency sensor technology for human–computer interaction applications, such as gesture recognition, and human activity recognition more broadly. While there is a significant amount of study on these topics, in most cases, experimental data are acquired in controlled settings by directing participants what motion to articulate. However, especially for communicative motions, such as sign language, such directed data sets do not accurately capture natural, in situ articulations. This results in a difference in the distribution of directed American Sign Language (ASL) versus natural ASL, which severely degrades natural sign language recognition in real‐world scenarios. To overcome these challenges and acquire more representative data for training deep models, the authors develop an interactive gaming environment, ChessSIGN, which records video and radar data of participants as they play the game without any external direction. The authors investigate various ways of generating synthetic samples from directed ASL data, but show that ultimately such data does not offer much improvement over just initialising using imagery from ImageNet. In contrast, an interactive learning paradigm is proposed by the authors in which model training is shown to improve as more and more natural ASL samples are acquired and augmented via synthetic samples generated from a physics‐aware generative adversarial network. The authors show that the proposed approach enables the recognition of natural ASL in a real‐world setting, achieving an accuracy of 69% for 29 ASL signs—a 60% improvement over conventional training with directed ASL data.

Respiration-related disease refers to a wide range of conditions, including influenza, pneumonia, asthma, sudden infant death syndrome (SIDS) and the latest outbreak, coronavirus disease 2019 (COVID-19), and many other respiration issues. However, real-time monitoring for the detection of respiratory disorders is currently lacking and needs to be improved. Realtime respiratory measures are necessary since unsupervised treatment of respiratory problems is the main contributor to the rising death rate. Thus, this paper reviewed the classification of the respiratory signal using two different approaches for real-time monitoring applications. This research explores machine learning and deep learning approaches to forecasting respiration conditions. Every consumption of these approaches has been discussed and reviewed. In addition, the current study is reviewed to identify critical directions for developing respiration real-time applications. [ABSTRACT FROM AUTHOR]

Similar waveforms and overlapping frequency ranges make distinguishing blasting vibrations and microseismic signals challenging, causing interference with coal mine microseismic monitoring systems. To address this problem, we propose a spectrum dataset (MSData) reflecting the spectrum features of both signal types and present a signal classification network (SCNet) combining CNNs and Transformers for signal classification. The network can learn multi-dimensional features of both signals from MSData and automatically and efficiently identify the two signal types. Experimental results yield F1-scores of 0.991 for microseismic signals and 0.993 for blasting vibration signals, meeting engineering application requirements. [ABSTRACT FROM AUTHOR]

Schizophrenia is a chronic mental illness that can negatively affect emotions, thoughts, social interaction, motor behavior, attention, and perception. Early diagnosis is still challenging and is based on the disease's symptoms. However, electroencephalography (EEG) signals yield incredibly detailed information about the activities and functions of the brain. In this study, a hybrid algorithm approach is proposed to improve the search performance of the marine predator algorithm (MPA) based on chaotic maps. For evaluating the performance of the proposed chaotic-based marine predator algorithm (CMPA), benchmark datasets are used. The results of the suggested variation method on the benchmarks show that the Sine Chaotic-based MPA (SCMPA) significantly outperforms the other MPA variants. The algorithm was verified using a public dataset consisting of 14 subjects. Moreover, the proposed SCMPA is essential for EEG electrode selection because it minimizes model complexity and selects the best representative features for providing optimal solutions. The extracted features for each subject were used in the decision tree (DT), random forest (RF), and extra tree (ET) methods. Performance measures showed that the proposed model was successful at differentiating schizophrenia patients (SZ) from healthy controls (HC). In the end, it was demonstrated that the feature selection technique SCMPA, which is the subject of this research, performs significantly better in regard to classification using EEG signals. [ABSTRACT FROM AUTHOR]

Electroencephalography (EEG) provides a wealth of physiological and psychological information. Decoding EEG signals enables machines to recognize brain activity, a crucial aspect in brain-computer interaction and medical rehabilitation. However, the non-stationarity and inter-individual variability of EEG signals pose challenges for existing EEG signal classification models to achieve the desired level of cross-subject generalization, limiting the practical applications of EEG-based brain computer interface systems. Unsupervised Domain Adaptation (UDA) aims to improve the model's generalization performance on the target domain by minimizing the discrepancies between the source and target domain data distributions. Many researchers treat different subjects as distinct domains and utilize Unsupervised Domain Adaptation (UDA) in transfer learning to guide the model for effective cross-subject EEG Classification. Strategies involve reducing distribution discrepancies between the source and target domains by minimizing well-designed discrepancy metrics or using an adversarial discriminator to capture domain-invariant features. However, neglecting category-discriminatory (abbreviated as discriminative) features leads to the limited effectiveness of these domain-level alignment methods. To tackle these issues, we propose the Discriminative Clustering Domain Adaptation Network (DCDAN), aimed at utilizing an unsupervised approach to construct discriminative information for facilitating Unsupervised Domain Adaptation (UDA) in achieving category-level feature alignment between the source and target domains. Specifically, we employ a clustering algorithm based on adversarial domain adaptation to associate pseudo-labels with target domain samples. Implementing a self-supervised training framework enables the model to acquire discriminative features. Moreover, we introduce the Pseudo Label-Common Spatial Pattern (PL-CSP) component, which integrates the prediction confidence of pseudo-labels from target domain samples with discriminative information from source domain samples through a covariance matrix weighting strategy based on sample confidence. This integration enhances the robustness of the learned spatial filter on the target domain. Experimental results on public datasets demonstrate the effectiveness of our proposed method in extracting more discriminative features. [ABSTRACT FROM AUTHOR]

The rigid guide is a crucial component of the mine hoisting system, which plays a role in guiding the smooth operation of the hoisting container in the process of mine hoisting. To address the issue of detection devices mounted on mobile equipment affecting normal production, this paper proposes to install the device inside the groove of the rigid guide, and directly collect the vibration signal of the rigid guide while the mine hoisting system is in operation. The collected vibration signals are preprocessed and subjected to fast Fourier transform. To fully extract the fault information hidden in the spectrogram, the vibration signals are transformed into a two-dimensional spectrogram in polar coordinates and used as a sample dataset for training a convolutional neural network (CNN) to achieve fault classification and identification of the rigid guide. Experimental studies on this method show that the accuracy of CNN in identifying rigid guide fault categories reaches 92.63%. Compared to the method of collecting vibration signals from mobile devices, the fault identification accuracy also exceeds 90%. By analyzing the vibration signals of the rigid guide, it is possible to determine whether there is a fault. [ABSTRACT FROM AUTHOR]