Signals doi: 10.3390/signals7050100

Authors: Yedukondala Rao Veeranki Hugo F. Posada-Quintero

Deep learning for biomedical signal analysis is revolutionizing the analysis of complex physiological signals through the extraction of intricate patterns from physiological and wearable sensor data. This Special Issue brings together five articles that address cardiac, neural, muscular, and motion signals using complementary approaches across signal processing, machine learning, deep representation learning, and wearable sensing. Two contributions focus on ECG: one introduces a constraint-based method to suppress false positives in P/R/T peak detection, and the other one presents an interactive GUI for rapid R-peak correction. In the third study, an EEG biometrics system was developed that uses 3D brain mapping and autoencoder features, and the fourth study compares machine classifiers for athlete movement recognition from EMG. In the final study, wearable IMUs were used to improve jump height estimation via optimal sensor placement. We summarize these works and highlight themes such as physiological prior integration, human–AI interaction, and multi-sensor fusion. This Editorial concludes with the identification of future directions, including multi-modal modeling and adaptive learning in real-world settings.

Signals doi: 10.3390/signals7050099

Authors: Selma Ozaydin

Comparing scale-parameter ratios in digital signal processing via unbounded logarithmic axes or periodic phase-based comparators introduces structural costs, including unbounded axis growth and cycle slips. We introduce GTFS, a bounded, nonperiodic angular projection mapping an unbounded ratio space onto a finite angular interval using a monotonically increasing, asymptotically saturating nonlinear function. Analytical properties, including monotonicity, boundedness, and octave-dependent angular contraction, are derived and verified across three closed-form realizations (arctangent, logistic, hyperbolic tangent). We examine four properties for multi-component signals: stateless order-independent projection, component-wise reliability-weighted selective alignment, bounded layer-consistent representation, and empirical robustness to cycle slips. Validated analytically, on synthetic multi-harmonic signals, and on 64 real LibriSpeech utterances, the projection achieves stateless independence to machine precision and reduces low-reliability component displacement up to 4.7-fold on real speech. The layered representation of a six-octave signal remains 19 to 41 times narrower than unlayered logarithmic mapping. In comparison against a classical phase-locked loop (PLL), GTFS exhibits zero cycle-slip events across all test conditions where the PLL fails. GTFS provides a bounded, memoryless, and cycle-slip-free alternative to conventional phase-tracking, with results validated on real acoustic data and properties expected to extend to vibration-based and communication signal-processing domains.

Signals doi: 10.3390/signals7050098

Authors: Alvertos Ioannis Mourikis Iosif Polenakis Christos Sarantidis Romanos Kalamatianos Ioannis Karydis

Smart agriculture is a key application domain of the Fourth Industrial Revolution (4IR), where cyber–physical systems, the Internet of Things (IoT), and data-driven analytics converge to improve productivity, sustainability, and resource efficiency. In this context, large volumes of sensor data collected in the field form the basis for automated monitoring, decision support, and adaptive control across agricultural processes. However, the value of such data depends not only on sensing technologies and connectivity, but also on how data are generated, transferred, processed, and interpreted throughout their lifecycle. This paper is organized in two parts. The first part reports a narrative review of the literature, based on a documented search protocol, and provides an overview of sensor data and sensor data transmission networks in smart agriculture. It examines the main challenges, potentials, and limitations associated with IoT-based field deployments in 4IR environments, and consolidates them into ten open problems. These problems are ordered from the device layer to the socio-economic layer and span physical robustness, device security, rural connectivity, aerial sensing, interoperability, labelled-data scarcity, explainability, federated access control, data ownership, and cost. The second part responds to these open problems with the conceptualization of a layered architecture and a set of design proposals, covering hierarchical edge security, offline-first synchronization, semantic interoperability, explainable decision support, and farmer-centric data governance, that are grounded in established techniques and standards but have not been implemented or empirically validated in this work; they are intended as a design blueprint and a starting point for future prototyping and evaluation rather than as measured results. The paper is accordingly positioned as a conceptual, perspective-oriented contribution: it states the procedure by which the open problems were derived from the reviewed works, it states what the proposed integration is claimed to add over existing smart-agriculture architectures, and it sets out an evaluation plan that fixes the metrics and the baselines against which each proposed component would have to be tested. Throughout both parts, particular attention is given to the role of data integrity and provenance as foundational elements for trustworthy agricultural data, enabling reliable analysis, auditability, and long-term reuse. By synthesizing current approaches, open issues, and conceptual solution patterns, this paper aims to support the design of resilient, transparent, and future-ready smart agriculture systems aligned with the principles of the 4IR.

Signals doi: 10.3390/signals7050097

Authors: Rongxiu Chen Linfeng Lian Zihao Deng Yuting Su Shuming Gao Linfeng Zheng

Accurate restoration and frequency tracking of non-stationary Doppler signals remain difficult under mixed and distribution-shifted noise. This study proposes EMD-THNet, a supervised one-dimensional framework that preserves the full 4096-sample resolution after offline empirical mode decomposition (EMD), combines a raw-waveform branch with a shared IMF encoder and energy-guided masked attention, and uses a hierarchical temporal encoder-decoder with waveform-restoration and instantaneous-frequency heads. The model was trained, validated, and tested using 96, 32, and 32 independently generated AM/FM windows sampled at 500 MHz. Robustness was further assessed through 1080 paired evaluations covering six input-SNR levels and six noise types. On the held-out synthetic set, EMD-THNet achieved an output SNR of 14.470 dB, an SNR gain of 12.296 dB (95% CI: 10.849–13.742 dB), and an SSIM of 0.8176. Six baselines were organized as classical filtering, Bayesian state estimation, and deep-learning temporal models. Relative to the strongest SNR-gain baseline in each group, EMD-THNet showed paired output-SNR advantages of 9.499, 4.676, and 2.389 dB, respectively (Holm-adjusted Wilcoxon p ≤ 7.68 × 10−8 for these three group-leading comparisons). On TSMS-Drone real I/Q waveforms with controlled added noise, zero-shot transfer yielded an SNR gain of 0.556 dB (95% CI: −0.316–1.416 dB), whereas adaptation using disjoint training distances yielded 10.241 dB (95% CI: 10.015–10.449 dB). These results demonstrate the effectiveness of EMD-THNet across the evaluated synthetic and semi-synthetic protocols and indicate its potential for robust Doppler-signal restoration and frequency tracking in future real-flight sensing applications.

Signals doi: 10.3390/signals7050096

Authors: Yanzhong Ma Chi Chen Lu Chen Ji Zhang Xiaojun Tian Guangrui Wen Zihao Lei

In the intelligent operation and maintenance of core process equipment in semiconductor manufacturing, the thickness and morphological parameters of sub-micron multilayer transparent films on wafer surfaces are critical quality indicators that govern device performance and yield. White Light Interferometry (WLI) enables the high-precision measurement of thin-film thickness and topography parameters, and is widely deployed in high-precision manufacturing fields such as semiconductors. However, when measuring sub-micron multilayer transparent films, WLI faces challenges including low computational efficiency, severe parameter coupling, and non-unique solutions, making it difficult to meet the demands of high-throughput online inspection. To address these issues, this paper proposes a deep learning-based method for decoupling and optimizing WLI signals through inverse modeling. The approach establishes an innovative hybrid intelligent framework: first, a training dataset is generated based on interferometric system modeling and simulation; then, a classification model is employed to intelligently categorize the acquired interference signals, decomposing the complex multimodal inversion problem into several sub-problems with simpler patterns. Next, for each signal category, a specialized closed-loop deep network model is designed and trained. This network integrates an inverse prediction network with a forward reconstruction network in series. During training, both prediction error and reconstruction error are jointly used as the loss function, ensuring solution uniqueness and physical consistency, thereby enhancing inversion reliability and robustness. This research provides an effective solution for high-precision online optical measurement of complex thin-film structures, offering significant theoretical value and broad industrial application prospects.

Signals doi: 10.3390/signals7050095

Authors: Nur Insyirah Iman Mohd Azman Tee Connie Ahmad Al-Khatib Mahmoud E. Farfoura

Parkinson’s disease (PD) causes motor control deficiencies, resulting in gait irregularities such as shorter strides, slower walking speed, and irregular step timing. This study presents a video-based deep learning approach for PD classification that extracts skeletal keypoints from Timed Up and Go (TUG) test videos using AlphaPose and the COCO-17 representation. A total of 24 features were generated, comprising 23 conventional gait features and one Freezing of Gait (FoG) feature derived from frequency-domain analysis of ankle velocity signals. This FoG feature was not validated against clinician-confirmed FoG episodes and should be interpreted as a frequency-domain proxy rather than a diagnostic measure. Three feature selection procedures and four LSTM-based architectures were evaluated across full, walking, and turning segments. Experimental results on a self-collected TUG dataset showed that the Standalone FI configuration achieved the numerically highest test accuracy of 77.78% on the turning segment among the evaluated LSTM configurations, while conventional features achieved 66.67% on both the full and walking segments. These test-set metrics provide descriptive estimates derived from a static subject-level test division involving six held-out participants (excluded from training and validation) and should not be viewed as statistically dependable indicators of clinical performance at the population level; in addition, gait-cycle boundaries were not independently validated and fallback usage was not quantified. Turning segments demonstrated higher discriminative power than straight-walking segments. Zero-shot cross-dataset evaluation on Turn-REMAP and PD-Walk revealed a substantial generalization gap, with accuracy falling to 55.12% and 49.53%, respectively, indicating that the present model is not yet suitable for cross-site clinical deployment without adaptation or calibration. The proposed framework provides systematic insights into the comparative role of conventional and FoG-derived gait parameters for non-invasive video-based PD screening.

Signals doi: 10.3390/signals7050094

Authors: Buma Xiao Huafeng He Liyuan Wang Yongquan You Yuancong Xiong

In scenarios where mainlobe interference and steering vector (SV) mismatch coexist, conventional robust adaptive beamforming techniques are prone to main-beam distortion, desired signal self-cancellation, and degradation of output performance. To address this problem, this paper proposes a mainlobe interference suppression method based on robust adaptive beamforming in the space-polarization joint domain. Firstly, the extended form of the Maximum Entropy Power Spectrum (MEPS) for polarization-sensitive arrays is theoretically derived. Owing to the high-resolution capability of MEPS under limited snapshots and multiple closely spaced mainlobe interferences, the Capon power spectrum is replaced by it for covariance matrix reconstruction. Then, a convex optimization problem with newly designed constraints is formulated to estimate the SV of the desired signal, which prevents the estimated SV from biasing toward interference directions. Subsequently, diagonal loading technology is introduced to further strengthen the robustness of the proposed method. Simulations are conducted under three typical SV mismatch scenarios. The results show that the proposed method exhibits stronger robustness under the coexistence of mainlobe interference and SV mismatch, achieving output performance closer to the theoretical optimum.

Signals doi: 10.3390/signals7050093

Authors: Yousheng Yang Lei Feng Yiding Liu Peng Xu Huaming Zhang Yumeng Sun Yonggang Xu Kun Zhang

Rolling bearing faults typically exhibit sideband structures in the frequency domain, yet conventional blind deconvolution methods operate in the time domain and rely on prior fault periods. To reduce this dependence and broaden applicability, this paper presents a frequency-domain blind deconvolution method based on weight feature extraction (WFE-FDBD). The method constructs rectangular pulse weights with a finite bandwidth to mitigate the effects of limited frequency resolution and minor sideband fluctuations. By incorporating the correlated kurtosis index, an inverse filter with adaptive period adjustment is built, enabling automatic selection of the fault period. Operating directly in the frequency domain, the proposed approach effectively enhances fault-related harmonics and suppresses noise interference. Both simulated and experimental signals validate the WFE-FDBD method, demonstrating its capability to diagnose localized defects on inner and outer raceways of rolling bearings.

Signals doi: 10.3390/signals7050092

Authors: Viktor Makarichev Vladimir Lukin Iryna Brysina

Large-scale image databases are continuously being accumulated across diverse information systems, and JPEG is the de facto standard for storing full-color images. We consider the problem of reducing the storage expenses of large volumes of photorealistic JPEG images with controlled distortion levels and without significant degradation in subsequent classification accuracy of convolutional neural networks. To address this problem, we apply the Discrete Atomic Compression (DAC) algorithm, which is based on atomic functions. Prior work has shown that this method combines compression with image content protection in a way that is relevant to modern secure image processing. For comparison, we also include a Progressive DCT-based Coder (PDCTC). Experiments are conducted on the ImageNet validation set (50,000 images across 1000 classes). Compression performance is assessed using MAD, RMS, PSNR, and compression ratio (CR). The compressed images are then classified by 28 pre-trained CNN classifiers spanning eight architectural families from established (VGG, ResNet) to state-of-the-art (EfficientNetV2) models. Classification is evaluated using Top-1 and Top-5 accuracy, among other metrics. DAC achieves a mean compression ratio of 1.93 relative to JPEG and PDCTC achieves 1.58. Both methods provide high reconstruction quality (mean PSNR > 36 dB). Across all classifiers, the Top-1 accuracy drop remains below 1.73 percentage points for DAC and 1.39 for PDCTC. Thus, DAC is promising for large-scale storage of JPEG images, offering stronger compression at the cost of a marginally larger classification accuracy drop compared to PDCTC.

Signals doi: 10.3390/signals7050091

Authors: Ehab El-Shazly Assem Abdelhakim Sherief Hashima

Neutron/gamma Pulse Shape Discrimination (PSD) is a critical task in radiation detection systems, where reliable classification becomes increasingly challenging under low Signal-to-Noise Ratio (SNR) conditions. Although Deep Convolutional Neural Networks (DCNNs) have demonstrated promising performance in automated signal classification, their robustness can degrade when detector signals are affected by varying noise levels. This paper proposes a Physics-Guided Feature Fusion with AlexNet (PGFF-AlexNet) framework that integrates time-frequency representations learned from scintillation pulses with complementary domain-specific physical descriptors. First, one-dimensional neutron and gamma pulses are transformed into two-dimensional Continuous Wavelet Transform (CWT) scalogram images to preserve their temporal and spectral characteristics. A comparative evaluation of four pretrained Convolutional Neural Network (CNN) architectures (AlexNet, GoogLeNet, EfficientNet-B0, and MobileNetV2) is then conducted over SNR levels ranging from 5 to 35 dB to identify a suitable visual feature extractor. Based on its overall discrimination capability and computational characteristics, AlexNet is selected as the visual backbone of the proposed framework. Its deep scalogram features are fused with a low-dimensional physics branch incorporating charge-comparison, frequency-domain, rise-time, and pulse-gradient descriptors. The resulting visual and physics-based representations are projected and concatenated prior to classification, enabling the network to exploit both learned time-frequency patterns and physically meaningful pulse-shape information. Experimental results show that PGFF-AlexNet achieves accuracies of 0.7369, 0.8900, and 0.9644 at 5, 10, and 15 dB, respectively, and an aggregate multi-SNR accuracy of 0.9548. The proposed framework provides substantial improvements over the standalone CNN baselines under challenging low-to-moderate SNR conditions and achieves an aggregate Matthews Correlation Coefficient (MCC) of 0.9090 and an Area Under the Receiver Operating Characteristic Curve (AUC) of 0.9912. Feature-space analyses using Principal Component Analysis (PCA), t-distributed Stochastic Neighbor Embedding (t-SNE), and intermediate activation maps further illustrate the complementary contribution of the physics-guided representation. These results demonstrate that PGFF-AlexNet provides an accurate, noise-robust, and interpretable framework for neutron/gamma PSD across diverse operating conditions.

Signals doi: 10.3390/signals7050090

Authors: Jarett Dewbury Chi-en Amy Tai Alexander Wong

Automated prostate cancer (PCa) lesion segmentation using deep learning remains constrained by limited tissue contrast in standard diffusion-based MRI sequences, with state-of-the-art methods reporting Dice scores of 32% or lower on large patient cohorts. Synthetic correlated diffusion imaging (CDIs) offers a promising solution, providing enhanced tissue contrast derived entirely from existing diffusion-weighted imaging (DWI) acquisitions at no additional clinical cost. This study presents the first comprehensive evaluation of CDIs integration across the full standard multiparametric MRI protocol, encompassing 15 modality configurations and six segmentation architectures spanning CNN and transformer families on a cohort of 200 patients. CDIs reliably enhances or preserves segmentation performance in the evaluated configurations, with 19 statistically significant improvements and no significant degradations across 42 direct comparisons. CDIs enhancement primarily operates as a recall-driven mechanism, improving lesion detection sensitivity while largely preserving precision. CDIs + DWI + T2w emerged as the strongest clinically meaningful configuration, achieving significant Dice improvement in four of six architectures, with no instances of degradation. Grad-CAM-based explainability analysis further reveals that CDIs focuses poorly localized CNN attention toward lesion boundaries, while transformer architectures exhibit more stable attention patterns that are less sensitive to CDIs integration. These results establish validated CDIs integration pathways and provide architecture-specific deployment guidance for clinical implementation.

Signals doi: 10.3390/signals7050089

Authors: Luella Marcos Paul Babyn Javad Alirezaie

Low-Dose Computed Tomography (LDCT) reduces radiation exposure but introduces severe quantum noise and streak artifacts that degrade image quality. To address these challenges, we propose the Physics-Guided Sequential State Space Transformer (PS3T), a projection-domain denoising framework that combines sequential state-space modeling with a photon-aware attention mechanism to capture long-range dependencies across projection angles with linear computational complexity. A differentiable Filtered Backprojection (FBP) layer further enforces reconstruction-domain consistency during training. The proposed framework was evaluated on the Mayo Clinic LDCT and Projection Dataset using patient-level dataset partitioning. Experimental results demonstrate that PS3T consistently outperforms state-of-the-art methods, including DRL, SADiff, and GEDFormer, across the abdomen, head, and chest datasets. On the abdomen dataset, PS3T reached a peak PSNR of 42.40 dB, an SSIM of 0.9020, and the lowest RMSE of 0.0076 across anatomical regions. Statistical analysis using 95% confidence intervals and paired Wilcoxon signed-rank tests confirmed that these improvements were significant (p<0.05). Furthermore, PS3T achieved the lowest reconstruction consistency loss (0.0128 at epoch 50), demonstrating stable convergence and the effectiveness of incorporating acquisition physics into projection-domain LDCT denoising.

Signals doi: 10.3390/signals7050088

Authors: Nurzhigit Smailov Maigul Zhekambayeva Dina Bauyrzhankyzy Gulbakhar Yussupova Alima Mambetaliyeva Aruzhan Nazarova Kuanysh Mussilimov Akezhan Sabibolda

Respiratory-sound event classification is challenged by non-stationarity, class imbalance, heterogeneous acquisition, and participant-correlated recordings. This study evaluates whether direct fusion of short-time Fourier transform (STFT), mel-frequency cepstral coefficient (MFCC), and wavelet-packet features improves a temporal one-dimensional convolutional neural network (1D-CNN), and whether temporal convolution offers an advantage over conventional classifiers. A radial basis function support vector machine (RBF-SVM) and random forest were included deliberately to separate the value of the engineered representation from classifier complexity. Experiments used 920 recordings and 6898 annotated cycles from the International Conference on Biomedical and Health Informatics (ICBHI) 2017 Respiratory Sound Database. The predefined 60:40 recording-level benchmark partition was retained; model selection used five-fold participant-grouped cross-validation, and 95% confidence intervals were estimated from 1000 participant-level bootstrap resamples. The complete hybrid 1D-CNN achieved a macro-averaged F1-score of 0.313. MFCC alone yielded 0.320, but the paired difference was not statistically resolved. The RBF-SVM and random forest achieved 0.401 and 0.378, respectively. These findings apply to direct early concatenation with the shared 1D-CNN backbone and do not imply that feature fusion is generally ineffective. The study provides leakage-aware baselines, controlled ablations, clustered uncertainty estimates, and frozen artifacts for reproducible comparison.

Signals doi: 10.3390/signals7050087

Authors: Takumi Morimoto Kazuhiko Kawabata Okana Hirota Meiko Ishikawa Nguyen Hai Chau Tomoyuki Yamamoto

In this study, the effect of data augmentation on machine learning (ML) models that predict the natural frequency of human long bones from the spectrogram of bone-conducted sound was investigated. Bone-conducted sound was recorded from the tibia of 45 healthy university students using a custom-developed hammering device. Short-Time Fourier Transform (STFT) was applied to the recorded sounds to generate spectrograms, from which six ML models were examined to predict the natural frequency. Ground-truth natural frequency was defined as peak values identified from Fast Fourier Transform (FFT) spectra obtained by recorded waveforms of sounds. Data augmentation was performed by shifting bone-conducted sound waveforms, expanded to 5 levels, i.e., 3-fold (3×) to 11-fold (11×). All models showed improved accuracy by data augmentations compared to the original dataset, with a particularly substantial improvement observed in many models when the dataset was expanded from the original to the 3× level. Among the ML models examined here, DenseNet-121 with 11× data augmentation achieved the lowest MAE and the highest R2 (10.00 ± 2.27 Hz; R2 = 0.927 ± 0.046), followed by PeleeNet, which is constructed on a lightweight DenseNet architecture, with 7× data augmentation (MAE = 10.37 ± 2.91 Hz, R2 = 0.920 ± 0.059).

Signals doi: 10.3390/signals7050086

Authors: Stefanos Balaskas

Multimodal Gazepoint Biometrics exports may combine eye-tracking, pupil, electrodermal, cardiovascular, engagement dial, TTL/event, and AOI information, but they are rarely analysis-ready. gpbiometrics provides a Gazepoint-aware R workflow v7.2.0 for importing heterogeneous files and folders, detecting schemas, determining time bases and active channels, auditing data quality, preparing signals, aligning events and AOIs, creating model-ready summaries, generating report-ready outputs, and handing data to specialist external toolchains. Its principal contribution is therefore not a new physiological estimator but an auditable vendor-aware layer between Gazepoint exports and established biosignal and statistical software. In the manuscript demonstration, the common synthetic audit executed 63 functions without error or duplication; the advanced audit reached 89 optional functions, of which 84 completed on the generic CSV demonstration, 4 required richer input objects, and 1 XDF/LSL helper was appropriately not exercised. The experimental cluster-permutation module is restricted to paired two-condition, one-dimensional participant-level time courses and is illustrated with a controlled known-effect simulation; unsupported advanced variants fail explicitly. Synthetic data are used to assess software execution and workflow structure, not physiological or behavioural validity. The revised manuscript also distinguishes core processing, lightweight approximations, external bridges, and guardrails and documents reproducibility and privacy-safe real-data smoke testing. gpbiometrics does not translate signal-derived outputs into direct claims about emotion, stress, cognition, preference, health status, diagnosis, mechanism, or exact temporal onset.

Signals doi: 10.3390/signals7050085

Authors: Xiao-Cong Zhong Deai Yin Xuefu Wang Qisong Wang Shiping Zhang

Electroencephalography (EEG)-based emotion recognition provides an important physiological basis for affective brain–computer interfaces. However, high-dimensional EEG features vary across trials, sessions, and subjects. Euclidean distance and conventional source-only LMNN cannot jointly learn a discriminative geometry and align cross-domain emotion distributions. To address this problem, we propose Adaptive Large Margin Nearest Neighbor (AdaLMNN), an unsupervised domain-adaptation framework for EEG emotion recognition. AdaLMNN learns a shared Mahalanobis projection through normalized large-margin discrimination, adaptive margins, and refreshed source neighborhoods. It further uses confidence-guided target pseudo-labels, class-balanced conditional alignment, and scale regularization with scheduled optimization to reduce source–target mismatch without using target ground-truth labels. We evaluated AdaLMNN on synthetic data and the SEED and SEED-IV datasets using SVM, KNN, and MLP under three protocols. AdaLMNN substantially improved EEG emotion-classification performance, particularly under within-session and session-level shifts. Although cross-subject gains remained dataset- and classifier-dependent, the results demonstrate the potential of AdaLMNN as an interpretable and extensible representation-learning module for EEG emotion recognition under heterogeneous recording conditions.

Signals doi: 10.3390/signals7040084

Authors: Zhiming Cheng Hongyu Tang Kai Wang Roujia Zhang Xiao Yuan

As the core power equipment of the deep-sea mining system, the deep-sea mineral lifting pump continuously operates under harsh service conditions coupled with high hydrostatic pressure, complex marine environments, and solid particle media. Its operating state shows obvious strong nonlinearity, time-varying fluctuation, and local abrupt change features. Therefore, time series health state prediction of the deep-sea mineral lifting pump is of vital engineering significance for realizing predictive maintenance and ensuring the safety of offshore trials and mining operations. Taking the 500 m-level offshore sea trial conducted in the Xisha area of the South China Sea as the engineering background, four critical health characteristic parameters, including shaft power, pump efficiency, motor winding temperature, and outlet radial vibration, are selected to construct a hybrid CNN-LSTM time series prediction model. Comprehensive model evaluation metrics and ablation comparison experiments are adopted to analyze the multi-step-ahead prediction performance of the proposed model. The results show that the CNN-LSTM model achieves optimal comprehensive evaluation indices in one-step prediction and possesses excellent tracking capability for inflection points and amplitude fluctuations of time series data. Although the prediction accuracy decreases gradually with the increase in prediction steps, the model can still effectively characterize the evolutionary trend of pump operating states, and its overall prediction performance is significantly superior to that of single models. This study provides model support and technical reference for the health evaluation, early fault alarm, and maintenance optimization of deep-sea mineral lifting pumps in offshore trials.

Signals doi: 10.3390/signals7040083

Authors: Ejay Nsugbe

Signal representation underpins modern signal processing, yet many existing methods primarily transform signals into alternative domains without explicitly modelling how informative signal structure evolves during recursive localisation. This paper presents the Linear Series Decomposition Learner (LSDL), a multi-geometric theory of signal structure and representation founded on recursive support localisation. The LSDL is formulated as a recursive localisation operator acting on a fixed amplitude-reference domain, thereby establishing a mathematically rigorous framework for analysing the evolution of signal support across successive localisation levels. Theoretical analysis characterises the fundamental properties of the operator, including support evolution, monotonicity, finite recursion, perturbation stability, and admissible localisation, and it thereby provides a formal foundation for recursive signal decomposition. Building upon this operator-theoretic formulation, the proposed framework establishes that recursive support localisation induces multiple complementary geometries of signal structure. These comprise support geometry, which describes the organisation of retained signal support; discriminative geometry, which characterises class separability under recursive localisation; information geometry, which quantifies entropy redistribution and information concentration; persistence geometry, which models the emergence, evolution, and lifetime of localised signal structures across recursive filtrations; and spectral geometry, which describes recursion-induced reorganisation within the frequency domain. Collectively, these complementary geometries provide a coherent multi-geometric representation that captures structural, statistical, topological, and spectral characteristics within a common mathematical framework. The proposed theory is supported through analytical development and empirical evaluation using synthetic benchmark signals, real-world electromyographic (EMG) datasets, and comparative analyses against established signal representation approaches, including the short-time Fourier transform (STFT), wavelet transforms, empirical mode decomposition (EMD), variational mode decomposition (VMD), and sparse coding. Experimental results demonstrate that recursive support localisation produces interpretable multi-geometric representations while maintaining competitive classification performance and low online computational cost. By establishing recursive support localisation as a principled mechanism through which complementary signal geometries emerge, the LSDL provides a mathematically grounded framework for interpretable signal representation, structural analysis, and representation learning across diverse signal-processing applications.

Signals doi: 10.3390/signals7040082

Authors: Fasikaw Kibrete Dereje Engida Woldemichael Hailu Shimels Gebremedhen Temesgen Tadesse Feisa Boaz Berhanu Tulu Orhan Çakar Erman Çelik

Fault diagnosis in industrial equipment plays a crucial role in ensuring reliable system functionality and minimizing the costs associated with repair and maintenance. Traditionally, fault diagnosis has relied on human intelligence (HI), with skilled personnel applying knowledge and expertise based on reasoning and contextual understanding. Nevertheless, as modern industrial systems have grown more complex, operated at higher speeds, and generated massive amounts of data, relying solely on HI has become increasingly challenging and less scalable. Consequently, modern fault diagnosis has turned toward artificial intelligence (AI). This paper presents a comparative study of HI and AI in industrial fault diagnosis, based on a literature-driven analysis. The results confirm that while AI-based fault diagnosis systems perform well in processing large datasets and achieve improved diagnostic accuracy, these practices also face limitations related to data dependency, explainability, and deployment cost. By contrast, human intelligence remains indispensable in handling uncertain, rare, or new fault conditions that require contextual judgment and flexibility. The review further indicates that augmented intelligence (AuI) provides a collaborative framework that combines the complementary strengths of HI and AI for industrial fault diagnosis. Furthermore, emerging research directions, such as explainable and trustworthy AI, foundation models, large language models, physics-informed AI, digital twins, and human-centered AI, are identified as promising developments for next-generation intelligent diagnostic systems. The findings suggest that augmented intelligence is the most promising approach for advancing the performance and reliability of diagnostic systems in industrial machines.

Signals doi: 10.3390/signals7040081

Authors: Shaoyuan Weng Zongwen Fan Liton Devnath

The liver plays a crucial role in maintaining essential physiological functions; however, excessive alcohol consumption significantly increases the risk of liver disorders. Accurate and early prediction of such conditions is vital for timely intervention and effective clinical management. Nevertheless, liver disorder prediction is typically challenged by class imbalance, which makes the conventional fixed classification threshold (0.5) suboptimal for binary classification. To address these issues, this paper proposes an AI-driven two-stage feature fusion-based ensemble model for intelligent biomedical data processing and liver disorder prediction. In the first stage, multiple tree-based ensemble models are employed to evaluate feature importance, and a feature selection strategy is designed to select an optimal subset of discriminative features. In the second stage, prediction probabilities generated by these base learners are integrated with the selected feature subset to construct an enhanced feature space through feature-level information fusion. This probability-aware fusion strategy captures richer predictive information and alleviates the limitations of fixed-threshold binary classification. In addition, a meta-ensemble model is employed to aggregate heterogeneous predictive patterns from multiple learners for the final prediction based on an optimized threshold. Extensive experiments based on two benchmark liver disorder datasets demonstrate that the proposed model consistently outperforms the compared models in terms of predictive performance. Statistical analysis also confirms that the proposed model significantly outperforms the compared methods. These results indicate that the proposed model could serve as a useful AI-aided biomedical healthcare data processing tool for liver disorder prediction.

Signals doi: 10.3390/signals7040080

Authors: Hazem Munaewer Al-Otum

In the contemporary digital era, image watermarking is essential for protecting intellectual property due to the widespread unauthorized distribution of digital content. In this work, a robust and efficient image watermarking scheme for copyright protection is proposed. The method integrates wavelet packet decomposition (WPD) with an ensemble of bagged tree classifiers, forming the BT-WPD framework. In the proposed approach, wavelet packet coefficients extracted from each color channel are reorganized into structured batches that capture spatial frequency characteristics, enabling effective watermark embedding in the WPD domain guided by the bagged tree ensemble model. Experimental results demonstrate that the proposed method achieves high imperceptibility, with a peak signal-to-noise ratio (PSNR) exceeding 60 dB, while maintaining strong robustness against various image processing attacks. The method also exhibits low computational complexity during watermark extraction, making it suitable for practical applications. Furthermore, the framework is extended to support Quick Response (QR) code watermark embedding, demonstrating enhanced robustness and versatility for copyright protection in digital media systems.

Signals doi: 10.3390/signals7040079

Authors: Matias Soto Ricardo Vergara Pablo Ormeño-Arriagada Jorge Vasquez

Field-based infrastructure inspection often occurs under intermittent or absence connectivity, limiting the applicability of cloud-dependent analytical systems. Existing approaches primarily focus on detection accuracy or edge processing, but rarely address the integration of reliable analytics, data traceability, and deployment constraints within a unified framework. To address this gap, we present OffInspect-LLM, an offline inspection platform integrating controlled LLM-assisted analytical interaction. The system combines an inspection pipeline for object detection with a constrained natural language interface that generates validated SQL queries, ensuring safe and traceable data interaction. Experimental evaluation on a dataset of 1600 images demonstrated stable multi-seed held-out test performance, achieving a deployment-oriented mean held-out test mAP@50:95 of 0.617 across five random seeds, while the highest exploratory single-run validation result reached 0.714 under fixed initialization conditions. The primary deployment-oriented evaluation corresponds to the multi-seed held-out test performance rather than the peak single-run validation result. In addition to predictive performance, the system achieves per-image inference times below 3 s on GPU, reliable batch processing, and scalable geospatial visualization exceeding 10,000 detections. The primary contribution lies in the integration of detection, structured data management, and audited querying within an offline-first architecture, enabling traceable and deployment-oriented inspection workflows through constrained analytical interaction under realistic operational conditions.

Signals doi: 10.3390/signals7040078

Authors: Shikha Chaudhary Gunjan Gupta Vikash Kumar Mishra Vipin Balyan Pramod Kumar Soni

Image steganography is an information-hiding technique, aiming to achieve confidentiality and data privacy while transmitting the data in a digital environment. Over the last two decades, steganography has evolved from classical spatial domain embedding to intelligent and adaptive steganographic systems capable of balancing imperceptibility, embedding capacity, robustness and security. This paper presents a review by categorizing the existing techniques into spatial domain-based, transform domain-based, hybrid and adaptive intelligent techniques. The review follows the PRISMA approach to make the selection process transparent for the inclusion and exclusion of papers in the study. Initially, the reviews include the spatial domain-based methods focusing on higher embedding capacity and simple embedding strategy, followed by transform-domain based techniques, including discrete cosine transform, discrete wavelet transform, and other multi-resolution wavelet transforms aiming to enhance robustness and imperceptibility by embedding the data into frequency coefficients. This paper further explores the methods that combine these techniques with other recent trends to develop adaptive and hybrid techniques. These techniques mainly integrate chaotic theory to enhance the security of secret data before embedding and optimization algorithms such as genetic algorithm, particle swarm optimization, Firefly, etc., for adaptive embedding to achieve an improved tradeoff. Finally, intelligent and adaptive techniques based on deep learning models such as convolutional neural networks, autoencoders, and generative adversarial networks are examined, highlighting their ability to learn intelligent embedding strategies and resist modern steganalysis. A comparative analysis is presented, including the technique, strengths, and limitations, together with the discussion of performance evaluation metrics and vulnerability analysis under image processing attacks. The review highlights the current trends and outlines the future direction to develop next-generation secure image steganographic systems.

Signals doi: 10.3390/signals7040077

Authors: Ji Won Kim Jae Yeong Kim Hanaro Park Soon-Hyun Ahn Eun-Jae Chung Jungirl Seok

Intraoperative monitoring systems play a vital role in surgical safety and decision-making. This study explored whether high-resolution physiological signals routinely available during surgery can be leveraged by deep learning to predict adverse events after head and neck free flap reconstruction. In this retrospective study, intraoperative waveforms, including arterial pressure, plethysmography, and electrocardiogram, from 187 patients who underwent free flap surgery were analyzed. A deep learning model based on the Mamba architecture was trained to predict three outcomes: flap failure, return to the operating room for exploration, and other surgical complications. Conventional logistic regression using static clinical variables and feature-based machine learning models were evaluated for comparison. On a patient-wise stratified held-out test set, the deep learning model achieved AUROCs of 0.86, 0.62, and 0.93 for flap failure, re-exploration, and other complications, respectively. Precision was 0.50, 0.40, and 1.00, whereas recall was 0.50, 0.40, and 0.25, indicating high precision but modest recall. Predictive performance varied across outcomes. These findings demonstrate the feasibility of waveform-based deep learning for perioperative risk stratification in reconstructive surgery, although the preliminary nature, limited sample size, and lack of external validation should be considered.

Signals doi: 10.3390/signals7040076

Authors: Ahmed El Badaoui Hicham Ben Alla Manal Hilali Said Ben Alla Abdellah Ezzati

Spatiotemporal deep learning models, like Graph Neural Networks (GNNs), Transformers, and selective State Space Models (Mamba), have achieved impressive performance in electroencephalogram (EEG) decoding and affective computing. However, their generalization performance often degrades severely under subject-independent Leave-One-Subject-Out (LOSO) cross-validation protocols. This generalization drop is often attributed to generic domain shifts in standard Brain–Computer Interface (BCI) literature, and is tackled by parameter-heavy adaptations. In contrast, this paper proposes a unified mathematical diagnostic framework to audit and measure the underlying representation and geometric collapse in spatiotemporal brain–computer interfaces. More concretely, we formalize: (1) Topological over-smoothing under volume conduction through Graph Dirichlet Energy bounds indicating GCNs as low-pass filters that smooth localized electrode variations; (2) representation collapse through the Normalized Rank Uniformity Index (NRUI) based on the Shannon Entropy of latent covariance eigenvalues, that distinguishes between dimensional and semantic collapse; and (3) geometric manifold distortions under subject domain shifts on the Symmetric Positive Definite (SPD) Riemannian manifold under the Affine-Invariant Riemannian Metric (AIRM) projection. Auditing these diagnostic metrics on canonical models across DEAP, DREAMER, and SEED, we demonstrate why standard spatiotemporal architectures suffer from performance collapse in cross-subject configurations. We provide BCI engineers with a tangible mathematical blueprint to design robust, collapse-resistant decoders.

Signals doi: 10.3390/signals7040075

Authors: Riaz-ul-haque Mian Michiko Inoue

Hardware–software co-design is a balanced strategy for computationally intensive algorithms such as decimal computing. It can provide several Pareto points for the development of embedded systems in terms of hardware cost and performance. In this study, we propose an efficient and accurate evaluation framework for decimal computing. The framework was designed and developed for hardware–software co-design decimal arithmetic using the RISC-V ecosystem. New binary and decimal-oriented instructions supported by an accelerator were developed. The framework can perform cycle-accurate analysis for performance and assess hardware overhead for co-design-based decimal arithmetic. Unlike previous studies that focused primarily on implementing and evaluating individual co-design methods, the proposed framework enables exhaustive hardware–software partition analysis at the building-block level, facilitating systematic exploration of the design space. We also evaluated the decimal floating-point multiplication Pareto points and identified a new Pareto point for hardware–software co-design-based decimal multiplication (Method-A) through an analysis with the proposed framework.

Signals doi: 10.3390/signals7040074

Authors: Abdelhakim Fares Abderrahim Fayçal Megri Abdallah Meraoumia

Biometric authentication systems are increasingly deployed in critical security applications, yet the irreplaceable nature of biometric traits poses fundamental risks when templates are compromised. Unlike passwords or tokens, biometric data cannot be reissued, making template protection a paramount concern for preserving both security and privacy. This paper presents a novel multilayer framework for biometric template protection that integrates cancellability directly into the feature extraction stage, ensuring non-invertible and revocable templates while maintaining high recognition accuracy. The proposed method employs chaotic projection of Binarized Statistical Image Features (BSIF) filter banks, optimized through Particle Swarm Optimization (PSO), to generate discriminative yet irreversible biometric templates. To strengthen security against statistical and cryptanalytic attacks, dual-layer scrambling and diffusion processes driven by chaotic maps eliminate spatial correlations and produce uniform intensity distributions. Furthermore, Random Grid Visual Cryptography (RGVC) divides the encrypted template into two shares stored in separate databases, ensuring that the compromise of a single repository reveals no biometric information. Extensive experiments conducted on the PolyU multispectral palmprint database demonstrate exceptional authentication performance, achieving Equal Error Rate (EER) values as low as 0.0520% after applying the proposed protection framework, under optimal configurations. Comprehensive empirical security analysis demonstrates favorable statistical security characteristics, including near-zero pixel correlation, near-uniform intensity distributions, high entropy values approaching the theoretical maximum of 8 bits, favorable NPCR and UACI values, and high sensitivity to key variations under the considered experimental settings. The proposed framework satisfies the essential requirements of cancellable biometrics, including diversity, revocability, and non-invertibility, while providing a privacy-preserving biometric template protection approach.

Signals doi: 10.3390/signals7040073

Authors: Riley Q. McNaboe Luís R. Mercado-Díaz Boluwatife E. Faremi Hugo F. Posada-Quintero

Electrodermal activity (EDA) has emerged as a promising physiological measure for objectively assessing neurodevelopmental disorders (NDDs) and related disorders, yet its effectiveness across conditions remains unclear. Following PRISMA guidelines, this review analyzed 23 studies (ASD: 11 studies, 410 participants; ADHD: 7 studies, 900 participants; OCD: 5 studies, 154 participants) published between 2010 and 2024 with a focus on methodology, including measurement protocols, signal processing, and analytical approaches. EDA demonstrated utility for diagnostic differentiation, predictive modeling, and physiological evaluation across conditions, with consistent patterns of heightened social arousal in ASD, medication-responsive sympathetic differences in ADHD, and impaired fear extinction in OCD. Machine learning approaches improved performance when combining EDA with multimodal physiological measures. However, substantial methodological heterogeneity existed across devices, recording sites, sampling frequencies, and signal processing, with 11 studies lacking signal processing details and 10 omitting sampling frequencies. The absence of standardized cross-disorder comparisons limited identification of disorder-specific versus shared autonomic signatures. While wearable technologies can enable continuous real-world monitoring, motion artifacts and signal quality remain challenges. Overall, standardized protocols and larger multimodal longitudinal studies are needed to establish EDA as a clinically useful biomarker for disorder assessment and personalized interventions.

Signals doi: 10.3390/signals7040072

Authors: Deepti C Annapurna Dammur

Background: Arrhythmia-induced cardiomyopathy (AIC) is a variation of cardiovascular disease with significant global morbidity and mortality. Arrhythmia is important to detect in time to allow for suitable intervention and clinical management, and this can be accomplished using electrocardiogram (ECG) signals. However, existing approaches face critical challenges to be solved, including (i) similar arrhythmias being hard to distinguish, (ii) misdiagnosis which can lead to life-threatening conditions, and (iii) single-lead ECG possibly missing subtle differences between arrhythmias. Methods: The study proposes a lightweight and efficient deep learning framework using selected ECG leads extracted from the PhysioNet large-scale 12-lead ECG arrhythmia database. The signals undergo preprocessing involving baseline wander removal and noise reduction. A Saliency-Guided Depthwise-Separable Dilated Network (SG-DSN) is introduced for feature extraction and classification. The model employs multi-scale dilated convolutions with saliency attention to capture discriminative ECG features, followed by softmax classification. The system not only classifies arrhythmia types but also flags potential AIC risk cases for early clinical intervention. Results: Experimental evaluation demonstrates improved arrhythmia classification performance, reduced computational complexity, and enhanced suitability for real-time clinical applications. Conclusion: Thus, the proposed framework is an efficient, scalable, and accurate solution for early detection and risk alerting of AIC.

Signals doi: 10.3390/signals7040071

Authors: Anastasios Papazafeiropoulos

This work analyzes the ergodic capacity behavior of reconfigurable intelligent surface (RIS)-assisted multiple-input multiple-output (MIMO) systems with a finite and arbitrary number of antennas and RIS elements under Nakagami-m fading conditions. By combining Hadamard’s determinant inequality with the Cauchy–Schwarz inequality, this work derives a dimensionally consistent closed-form upper bound on the ergodic capacity in terms of the Meijer G-function. Subsequently, it is demonstrated that at a high signal-to-noise ratio (SNR), a simplified expression for the capacity upper bound can be derived, enabling an analytical assessment of how the fading parameter influences the ergodic capacity. The study also explores the asymptotic behavior in the large-system regime, where the number of antennas or RIS elements tends to infinity. Monte Carlo (MC) simulations confirm the accuracy of the proposed bound and scaling laws.

Signals doi: 10.3390/signals7040070

Authors: Theodoros Saganas Spyridon Votis Fotios Kantas Foivos S. Kanellos Georgios Rigas Andreas P. Katsenos Yannis V. Simos Lampros Lakkas Georgios S. Markopoulos Vasiliki Kostadima Spyridon Konitsiotis Dimitrios Peschos Konstantinos I. Tsamis

Spasticity is a common and disabling consequence of upper motor neuron lesions, and its follow-up relies largely on clinical scales that may not fully reflect daily-life mobility. This prospective single-arm exploratory pilot study examined wearable-derived daily-life mobility metrics around a scheduled botulinum neurotoxin type A (BoNT-A) injection cycle and assessed their associations with established clinical measures. Thirteen patients with focal spasticity secondary to stroke (n = 11) or multiple sclerosis (n = 2) underwent baseline and 6-week follow-up assessment using the Modified Ashworth Scale (MAS), Motricity Index (MI), the Timed Up and Go test (TUG) and a multi-sensor wearable monitoring system. Device-derived outcomes included gait impairment, gait speed, stride length, angular velocity, and lack of movement. Clinical scales generally changed in the expected post-treatment direction, with reduced MAS scores and descriptive or directional improvements in MI scores and TUG performance. Wearable metrics captured descriptive changes in daily mobility, notably a reduction in lack of movement. Furthermore, in exploratory pooled analyses, selected gait-related metrics tended to be associated with TUG performance and lower-extremity MAS. These findings highlight the potential of wearable telemonitoring to complement conventional clinical scales by providing objective, real-world data, supporting its further validation as a tool for longitudinal spasticity management and neurorehabilitation. These device-derived metrics should be interpreted as mobility-related digital biomarkers rather than direct surrogate measures of spasticity.

Signals doi: 10.3390/signals7040069

Authors: Marcin Kołodziej Andrzej Majkowski Tomasz Rywik

This study compares the effectiveness of two approaches to speech emotion recognition for three affective states in Polish: sad, neutral, and happy. Both a set of acoustic features—capturing prosodic, phonatory, temporal, spectral, and cepstral properties—and representations learned by self-supervised models (wav2vec 2.0 and WavLM) were analyzed. Experiments were conducted on the nEMO corpus, comprising 2327 recordings from nine speakers, using a rigorous leave-one-subject-out protocol to evaluate cross-speaker generalization. In the feature-based approach, 107 acoustic features were used, and classification was performed with logistic regression and, additionally, SVM variants. In the deep learning approach, the wav2vec2-base and WavLM-base models were fine-tuned for the three-class task. The best results were achieved by the self-supervised models: WavLM reached a global balanced accuracy of 0.727 and a macro-F1 score of 0.710, while wav2vec 2.0 achieved 0.722 and 0.695, respectively. Both outperformed the feature-based approach (BAcc = 0.627, macro-F1 = 0.584). Confusion matrix analysis showed that the greatest difficulty lies in distinguishing the neutral class from the sad and happy classes, whereas sad and happy classes are more clearly separable. Feature utility analysis (SFS under the LOSO protocol) indicated the significant role of cepstral features (MFCCs and their derivatives), complemented by selected prosodic and temporal features. An additional comparison of SVM classifiers suggested that the main limitation of this approach lies in the signal representation itself rather than solely in the choice of classifier. Explainability analyses of the deep models, using layer-wise probing and integrated gradients, showed that affective information is best represented in intermediate layers, and that model decisions rely on locally salient segments of the signal. Furthermore, a speaker adaptation experiment demonstrated that personalization significantly improves classification performance, highlighting the potential of such methods for long-term monitoring of affective expression changes in the same individual.

Signals doi: 10.3390/signals7040068

Authors: Apsorn Sattayakhom Kosin Kalarat Waluka Amaek Pavarud Puangsri Matina Ngodngamthaweesuk Phanit Koomhin

Arousal is essential for cognitive awareness, and it typically declines during prolonged tasks. Currently, mobile devices are frequently used during breaks to relax and counteract this decline. However, their impact on neurophysiological recovery remains poorly understood. Therefore, this study compared the effects of traditional quiet rest versus a mobile task break on cortical arousal using spectral analysis of electroencephalography (EEG) signals in twenty healthy young females (20–25 years). Raw EEG data were transformed using Fast Fourier Transform (FFT) to determine power spectral densities, with a state of physiological underarousal first induced via a prolonged eyes-closed condition. Results revealed that this state was characterized by reduced alpha/beta power and delta/theta synchronization starting at the 4th minute. Although traditional quiet rest suppressed delta/theta synchronization, it failed to sustain cortical arousal, with alpha and beta powers declining by the 8th minute. In contrast, passive social media browsing acted as a potent neurocognitive stimulant, not only sustaining arousal but markedly increasing high-frequency beta power by the 16th minute. Furthermore, preliminary network-level connectivity analysis using Phase Locking Value (PLV) revealed that mobile tasks induced widespread beta-band synchronization across frontal-midline regions, suggesting enhanced functional coupling within the executive control network. In conclusion, in healthy young females, while mobile tasks strategically counteract low arousal, they fail to facilitate the neurophysiological disengagement necessary for true recovery. These findings underscore the importance of digital hygiene, highlighting a distinction between alertness-boosting activities and recovery-focused rest. The results suggest that mobile tasks may create a subjective perception of rest despite objective signs of sustained cortical activation, implying that such activities may not facilitate genuine neurophysiological recovery within the context of short-term neurophysiological modulation.

Signals doi: 10.3390/signals7040067

Authors: Myoungbeom Chung

Playground safety for preschool and early elementary school children has become an increasingly important issue for families and local communities. In semi-open playground environments, direct supervision by caregivers is often difficult, while conventional positioning approaches may be costly, infrastructure-dependent, or unreliable due to frequent signal obstructions. This study presents an integrated low-cost monitoring system that combines commercial BLE beacons, a single smart device installed in the playground, a push-notification server, and a caregiver smartphone application to detect children’s entrance and exit events and visualize their playground usage logs. Rather than proposing a new localization algorithm, the main contribution of this work is the practical system integration and field validation of BLE-based entrance/exit detection in semi-open playground settings. The smart device continuously scans beacon RSSI values, applies Kalman-filter-based smoothing and threshold-based state transitions, and transmits detected events to the server, where daily, weekly, and monthly statistics are generated and visualized for caregivers. Field experiments conducted at five real playgrounds showed that the proposed system achieved over 99% entrance/exit detection accuracy with an average response time of less than 7 s. These results demonstrate that reliable playground entry/exit monitoring can be implemented at low cost, with simple infrastructure and practical deployment in residential environments.

Signals doi: 10.3390/signals7040066

Authors: Devi Sowjanya Padala Lin Li Hetali Tank Ming Zhang Jeffrey B. Driban Timothy McAlindon Juan Shan

Segmentation of bone marrow lesions (BML) is vital for quantifying lesion volume, a biomarker associated with cartilage damage and pain in knee osteoarthritis (KOA). Manual segmentation is challenging due to low contrast and variable lesion locations. We propose an automated deep learning method featuring a dual-stream encoder that integrates global image-level and local patch-level features. The study included 300 participants from the Osteoarthritis Initiative (OAI) database, each with approximately 36 intermediate-weighted fat-suppressed (IWFS) magnetic resonance (MR) images. The ground truth masks were manually annotated by trained research staff. With physical batch size 32, the model achieved a 2D Dice similarity coefficient (DSC) of 0.68, 3D DSC of 0.62, Intersection over Union (IoU) of 0.51, precision of 0.76, sensitivity of 0.62, and Pearson’s correlation coefficient (r) of 0.85 between manually labelled and automatically generated volumes. Using an effective batch size of 64 via gradient accumulation, the model achieved 2D DSC of 0.63, 3D DSC of 0.65, IoU of 0.48, precision of 0.75, sensitivity of 0.6, and r of 0.98 for volume correlation. The model outperformed baselines at batch size 32 across almost all evaluated metrics and remained robust at batch size 64, with strong volumetric correlation and improved 3D DSC, IoU, and sensitivity.

Signals doi: 10.3390/signals7040065

Authors: Ali Sarafnia M. Omair Ahmad M. N. S. Swamy

The conventional method of optimizing the parameter of a first-order differential microphone array (DMA) in the presence of noise is to maximize the array gain, i.e., to maximize the noise reduction. Such optimization can be accomplished only in the case of channel noise scenario, where the maximum array gain exists. However, in the case of diffuse or point source noise, the array gain is not upper bounded when the microphones are very closely spaced. Hence, the classical method fails in these two cases. In this paper, we present a method of designing a first-order DMA that ensures effective noise reduction in all the three cases, namely, channel, diffuse, and point-source noise, by utilizing the concept of spectral entropy measure (SEM). Thus, the SEM-based method is superior to the existing maximum array gain method, since it provides the optimal design parameter in all the three noise cases. The performance of the first-order DMA designed using the spectral entropy-based measure is evaluated for an input speech signal in the presence of the three noise scenarios mentioned above. This research underscores the effectiveness of the spectral entropy-based measure in overcoming the limitations posed by the classical method of designing a first-order DMA.

Signals doi: 10.3390/signals7040064

Authors: Reka Sandaruwan Gallena Watthage Anil Fernando

Adaptive bitrate streaming over HTTP relies on carefully constructed bitrate ladders and ordered sets of bitrate–resolution pairs to deliver optimal perceptual quality under fluctuating network conditions. While content-aware methods based on convex hull optimisation have substantially improved ladder efficiency for Video-on-Demand, they require exhaustive multi-resolution pre-encoding that is computationally prohibitive under the real-time constraints of live streaming. This challenge is compounded by the H.266/Versatile Video Coding (VVC) standard, which offers approximately 50% compression gains over HEVC at 8–10× the encoding complexity. This paper presents LiveCH-VVC, a latency-aware dynamic bitrate ladder prediction framework for VVC-encoded live streaming over Low-Latency DASH (LL-DASH) with CMAF packaging. The framework introduces four integrated modules: (i) a Lightweight Dual-Path CNN (LDP-CNN), obtained via teacher–student knowledge distillation (∼5 M parameters, 148 ms GPU inference), that jointly extracts spatial–temporal features from raw frames and compression-domain statistics from a fast VVC probe encode; (ii) an adaptive scene change detector with exponential moving average thresholding (F1 = 0.925) that triggers ladder updates only upon significant complexity shifts; (iii) a temporally augmented XGBoost multi-label classifier that predicts latency-constrained Pareto-optimal bitrate–resolution pairs; and (iv) an online adaptation engine that integrates Common Media Client Data (CMCD) feedback from CDN edge servers for continuous closed-loop refinement. Comprehensive evaluation on 81 UHD sequences (∼4050 CMAF segments) from three benchmark datasets demonstrates an average BD-Rate of +0.68% relative to the per-segment oracle convex hull 5.4× better than the state-of-the-art ARTEMIS framework (+3.67%) while achieving 73.3% encoding time savings, 2.37 s end-to-end latency, and a QoE score of 81.6 in live simulation with 100 concurrent clients. Ablation analysis confirms that the dual-path compression-domain branch (+0.44 pp) and temporal context augmentation (+0.35 pp) are the primary performance drivers, while the online adaptation mechanism provides 42% relative improvement over extended streaming sessions.

Signals doi: 10.3390/signals7040063

Authors: Mingjie Qiu Jianming Wang Guangxin Wu

Aiming at core bottlenecks in weak and small target detection in radar range-Doppler (RD) spectra under low signal-to-noise ratio (SNR)—including severe performance degradation of traditional constant false alarm rate (CFAR) detectors and the inherent trade-off difficulty faced by existing deep learning methods in balancing detection accuracy, localization precision, and real-time performance—this paper proposes a progressive sub-pixel-level intelligent detection algorithm named F2DN-CCWL. The algorithm constructs a three-stage detection pipeline: global candidate screening, local fine discrimination, and weighted localization, and implements a full-stack customized design covering network architecture, soft-label training strategy, and post-processing modules. Simulation and field-measured results demonstrate that at −20 dB SNR, the proposed algorithm achieves a detection probability of 95.3%, a false alarm rate of 3.1%, an average localization error of 0.76 pixels, and a single-frame inference latency of 47.21 ms. This method offers a high-performance engineering solution for radar-based detection of low observable targets.

Signals doi: 10.3390/signals7040062

Authors: Yutaka Yoshida Kiyoko Yokoyama

Accurate R-peak detection in electrocardiogram (ECG) signals is fundamental for cardiovascular analysis. However, most existing methods address differences in sampling frequency (fs) through signal resampling or transfer learning, which may alter the temporal definition of annotated events. In this study, we propose a fs consistent framework for ECG R-peak detection that avoids both resampling and retraining. The proposed method is based on low-sampling morphological learning combined with physiological temporal constraints (PTC). A lightweight classifier based on Extreme Gradient Boosting (XGB) was trained on 128-Hz ECG data from the MIT-BIH Normal Sinus Rhythm Database to learn local morphological structures, and feature extraction is defined in milliseconds with time-normalized derivatives to ensure consistency across fs. The trained model is directly applied to higher-fs datasets (360 Hz, 500 Hz, and 1000 Hz) without modification. Final peak locations are determined through deterministic processing, including PTC and local snap processing. Experimental results demonstrated that the proposed method achieved stable detection performance across multiple sampling frequencies. When evaluated in a sample-wise manner, the proposed method achieved mean F1-scores of 0.885 on MIT-BIH Arrhythmia Database (360 Hz), 0.848 on Lobachevsky University Electrocardiography Database (LUDB, 500 Hz, sinus rhythm), 0.837 on LUDB (500 Hz, arrhythmia), and 0.953 on PTB Diagnostic ECG Database (1000 Hz), without any resampling or retraining. The integration of probabilistic candidate detection and deterministic temporal alignment enables consistent peak localization under cross-frequency conditions. These findings demonstrate that augmenting machine learning with deterministic decision mechanisms provides a principled framework for fs-consistent ECG peak detection.

Signals doi: 10.3390/signals7040061

Authors: Wejdan Altawallbeh Huthaifa Obeidat Issam Trrad Hazem Al-Otum

Photovoltaic systems have recently attracted significant attention for the free, clean, and sustainable energy they generate. In this work, thermal image processing techniques were developed and utilized to classify hot spots on solar photovoltaic panels. Thermal images were classified into three categories: (a) ideal images, where images do not contain hot spots; (b) images affected by shadow; and (c) images affected by bird drops. The proposed classification was developed using image processing techniques, including histogram analysis, contrast enhancement, and filtering tools. The attained classes are then matched to the decrease in electrical power output. The proposed method was applied to thermal images to detect and classify the target hot spot. Experimental results showed that the estimated error was approximately 6.3% of the total number of images used in the research, with error rates of 6.57% for the shadow hot spot type and 6.67% for the bird drops (mud-like class). Moreover, the accuracy of the proposed method was around 93.7%.

Signals doi: 10.3390/signals7040060

Authors: Hao Liu Jaehyeon Nam Jaecheon Lee Shunming Li Haibo Zhang Jiantao Lu Changpeng Cai

Intelligent fault diagnosis of rotating machinery typically assumes that training and test data share the same operating speed, an assumption that rarely holds in industrial end-of-line testing, where a rotor must be certified across a range of shaft speeds. In this paper, we expose this assumption through a systematic benchmark of four deep learning architectures (TCN, 1D-CNN, BiLSTM, and CNN-BiLSTM) on a laboratory rotor testbench with three operating speeds (1000, 2000, and 3000 rpm) and four unbalance fault classes. Under within-speed 5-fold cross-validation, all four models achieve a perfect macro-F1 of 1.000, offering no basis for architecture selection. Under Leave-One-Speed-Out (LOSO) evaluation (train on two speeds, test on the held-out speed), performance drops substantially and diverges across models: BiLSTM 0.180, TCN 0.270, 1D-CNN 0.271, and CNN-BiLSTM 0.401. We trace the LOSO gap to the unbalance centrifugal force law F = meω2, which makes speed-confounded features unreliable under cross-speed testing. CNN-BiLSTM improves the mean LOSO macro-F1 by 48% relative to the stronger single-module baseline, 1D-CNN. Although CNN-BiLSTM achieves the highest LOSO performance among the evaluated architectures, it still does not surpass the physics-informed LightGBM baseline of 0.487. Therefore, the primary contribution of this work is not to solve cross-speed diagnosis, but to demonstrate that conventional same-speed evaluation substantially overestimates model capability and that LOSO provides a more deployment-relevant benchmark for future algorithm development.

Signals doi: 10.3390/signals7040059

Authors: Marko Jurišić Igor Tomičić Andrija Bernik

User Behaviour Analytics (UBA) relies heavily on sequential data to detect anomalies such as insider threats. Traditional approaches often model user behaviour as discrete sequences of events using Recurrent Neural Networks (RNNs) like Long Short-Term Memory (LSTM) networks. These methods implicitly treat time steps as uniform, ignoring the irregular time intervals inherent in user logs. In this paper, we present the first application of Neural Controlled Differential Equations (NCDEs) to user behaviour analytics, a class of continuous-time models that naturally handle irregularly-timed event data. We compare a simple LSTM predictor against an NCDE predictor on the CERT 4.2 and 6.2 insider threat dataset. We demonstrate that standard discrete-time models (LSTMs) produce noisy loss signals on sparse data, forcing downstream classifiers to rely on fragile error spikes. In contrast, Neural CDEs generate stable, continuous error signals. NCDE roughly tripled the F1 of the discrete baseline (0.364 vs. 0.133) on the challenging CERT 6.2 dataset.

Signals doi: 10.3390/signals7030058

Authors: Nelson Chambi Celso Sanga Alejandra Sanga Piero Sanga

Predictive maintenance in underground mining faces challenges due to severe conditions such as confined environments, high humidity, presence of silica dust, and restricted access. This study develops a predictive framework based on oil analysis and machine learning for multiple compartments of mining equipment (engine, hydraulic system, transmission, differential). Samples were processed under ASTM standards, integrating wear metal concentrations (Fe, Cu, Cr, Pb, Al), physicochemical properties (viscosity, TBN, soot), and contaminants (Si, Na). Based on tribology, interpretable ratios were constructed. Three algorithms (Random Forest, Gradient Boosting, and XGBoost) were evaluated using cross-validation. XGBoost achieved the best balance (F1 = 0.852, AUC = 0.975), with a recall of 94.5% for the critical class and only 3 false negatives out of 199 test samples, while Random Forest presented the highest global discrimination power (AUC = 0.978). SHAP revealed that viscosity at 100 °C is the most important predictor (SHAP ~0.9), surpassing iron. No temporal wear trend was found (R2 = 0.000). Threshold optimization to 0.25 reduced false negatives by 67% (from 9 to 3). The framework provides interpretable predictions with uncertainty quantification for underground environments.

Signals doi: 10.3390/signals7030057

Authors: Salihu Dausu Ibrahim Emmanuel Majiyebo Eronu Aliyu Ozovehe Sanni Muhammad Uthman Sunday Oladayo Oladejo

Unauthorised FM broadcasting poses significant challenges to spectrum regulators globally, contributing to interference, degraded service quality, and national security threats. While traditional spectrum monitoring relies primarily on carrier frequency and power measurements, this study demonstrates that FM baseband features—specifically the multiplex (MPX) signal structure, pilot tone, and Radio Data System (RDS) subcarrier—provide robust discriminative markers for detecting non-compliant transmissions. Using a real-world dataset of 3710 pre-processed records collected across Nigeria’s capital region between 2021 and 2024, we extracted and analysed six transmission parameters: assigned frequency, band occupancy (±100 kHz), MPX overshoot percentage, pilot tone presence, and RDS indicators. A Support Vector Machine (SVM) classifier with radial basis function (RBF) kernel was trained to distinguish compliant licensed stations from regulatory non-compliant transmissions—encompassing both unlicensed transmitters and technically non-compliant licensed operators—achieving 99.96% accuracy, 99.38% precision, and 99.63% recall with a false alarm rate of 0.026%. A Comparative analysis against baseline feature sets confirmed that integrating MPX, pilot, and RDS significantly improved detection robustness compared with carrier-only approaches. Results demonstrate that feature-rich baseband analysis enables scalable, cost-effective regulatory enforcement, reducing manual monitoring burden while enhancing detection reliability. This framework offers practical applicability for spectrum management agencies in resource-constrained environments where unauthorised broadcasting remains prevalent.

Signals doi: 10.3390/signals7030056

Authors: Barlian Henryranu Prasetio Edita Rosana Widasari

Speech-based stress analysis is relevant to forensic-oriented speech processing, security screening, and behavioral monitoring, yet its reliability is often limited by speaker variability, recording conditions, and acoustic mismatch. This study proposes an interpretable multi-domain signal processing framework that models stress-related speech variation through excitation dynamics, vocal tract characteristics, and temporal modulation patterns. The framework integrates source–filter decomposition, residual-domain analysis, harmonic structure analysis, modulation spectrum characterization, and prosodic variability into a unified representation. The SUSAS corpus is used as the primary dataset for supervised stress evaluation. RAVDESS and SAVEE are employed only as controlled arousal-related proxy datasets to examine the consistency of stress-related acoustic patterns, rather than as physiological stress ground truth. VoxCeleb is used exclusively for robustness and domain-variability analysis because it lacks stress labels. For probabilistic evidence assessment, Gaussian mixture models are adopted as the more interpretable density estimator, while normalizing flow is included as a flexible performance-oriented comparator for modeling non-Gaussian feature distributions. Evaluation incorporates likelihood ratio analysis, DET curves, EER, ablation studies, and robustness testing. The proposed framework achieves an EER of 5.8% in the primary supervised evaluation, showing competitive performance while preserving physically meaningful interpretation.

Signals doi: 10.3390/signals7030055

Authors: Artem Bykov Nurkamilya Daurenbayeva Syrym Zhakypbekov Aigul Bissarinova Almas Nurlanuly Duriya Daniyarova

Distributed seismic sensor networks integrated into the Internet of Things (IoT) infrastructure enable continuous condition monitoring of large-scale engineering structures. During long-term operation, however, measurement channels are subject to sensitivity drift, increased noise, and pulse artifacts that statistically mimic real vibration events. Related deep-learning techniques for noisy and ill-posed inverse problems have demonstrated the value of combining principled physical priors with deep models. Although the application domain differs, the underlying methodological insight—that constrained, physics-aware feature mappings can stabilize learning under noisy and partially observed conditions—directly motivates the use of a parameterized quantum circuit as a nonlinear feature transformer in the present work, where Hilbert space mapping serves as an analogous structural prior for the latent representation. Three principal fault modes are considered in this work, corresponding to the dominant degradation mechanisms observed in long-term seismic instrumentation: sensor drift, increased noise, and sensor failure. Each fault mode produces a distinct signature in the windowed feature space; the proposed model is trained to discriminate between them based on the latent CNN-LSTM-VQC representation. We propose a hybrid quantum-inspired deep-learning model (QC-DL) for the detection and diagnosis of channel-degradation anomalies. The architecture combines a 1D-CNN+LSTM feature extractor with a parameterized variational quantum circuit (VQC) used as a nonlinear feature transformer. All quantum experiments were performed on the QPanda3 CPUQVM simulator. The data were split chronologically prior to windowing to avoid information leakage. On real-world labeled accelerometric data with four operating modes (normal/drift/high-noise/failure), the QC-DL model achieved a macro-averaged F1 score of approximately 0.69 and per-class AUC values in the range 0.88–0.99. The mean early-detection latency was 1.6 s versus 2.1 s for the CNN-LSTM baseline (~24% reduction). An ablation study against a parameter-matched classical MLP showed that the gain is modest and not solely attributable to additional nonlinearity. The reported p-values (p = 0.70, p = 0.29) do not establish statistical significance. The results support the feasibility of hybrid quantum-inspired deep learning for sensor-channel verification, while highlighting the need for evaluation on real NISQ hardware. This paper proposes a hybrid quantum-inspired approach for detecting and diagnosing such anomalies in the time series of distributed seismic networks. The architecture combines a classical temporal feature extraction module based on one-dimensional convolutional layers and a recurrent long short-term memory (LSTM) network, which generates a latent window representation of the signal, with a parameterized variational quantum circuit used as a nonlinear feature processor in a hybrid computational circuit. Experimental validation was performed on real-world labeled data with multiple sensor degradation modes. The evaluation was organized in a scoring framework aligned with autonomous operation through window ranking and threshold alarm generation. In the experiments, the proposed model provided a macro-averaged F1 score of approximately 0.69 and area under the receiver operating characteristic (AUC) curve values in the range of 0.88–0.99 across classes, outperforming baseline deep models. The average early detection latency was 1.6 s versus 2.1 s for the baseline recurrent model (a 24% reduction). An ablative comparison with a control model based on a classical multilayer perceptron of comparable dimension confirmed that the improvement is not limited to the addition of additional nonlinearity. The obtained results indicate the potential of quantum-supported deep learning for improving the reliability of long-term vibration monitoring and verifying the correctness of sensor channels in distributed seismic networks.

Signals doi: 10.3390/signals7030054

Authors: Luis Viloria Edmanuel Cruz Cesar Pinzon-Acosta

Manual arc welding is highly dependent on operator skill, leading to variability in weld quality and an increased risk of defects; therefore, reliable monitoring methods for Shielded Metal Arc Welding (SMAW) are required, particularly in manual environments where process variability and environmental noise are inherent. This study proposes a monitoring approach for classifying SMAW process parameters using airborne acoustic signals generated by the welding arc. Welding experiments were conducted on carbon steel plates of different thicknesses (3, 6, and 12 mm) using E6010, E6011, E6013, and E7018 electrodes under Alternating Current (AC) and Direct Current (DC) configurations; acoustic signals were recorded in real time and processed using Mel-Frequency Cepstral Coefficients (MFCCs) and deep audio embeddings from pre-trained VGGish and YAMNet models as inputs to artificial neural network classifiers for multi-output classification of welding process parameters. Model performance was evaluated using per-target metrics (accuracy and macro F1-score) and joint multi-output metrics (Exact Match and Hamming Accuracy). MFCC-based models significantly outperformed embedding-based approaches, achieving up to 94.51% Exact Match and 97.88% Hamming Accuracy, while reducing computational costs. These results demonstrate the feasibility of SMAW monitoring using arc sound, suggesting that spectral features are an effective solution for welding-process monitoring and a promising foundation for future weld-quality monitoring systems.

Signals doi: 10.3390/signals7030053

Authors: Weiquan Wang K.M Shahriar Alam Adib Foisal Ahmed Riaz-ul-haque Mian

Wafer and FPGA testing remains costly in semiconductor manufacturing. This paper studies random sampling, stratified sampling, and k-means sampling under a partial-measurement setting with Gaussian Process Regression (GPR), and introduces Short Distance Elimination (SDE), a spatial screening rule that spreads selected training points over the layout. Combining value-based sampling with SDE yields two hybrid methods: S-SDE, which applies SDE within stratified subsets, and K-SDE, which applies SDE within k-means clusters. A calibration-based protocol fixes the value-group labels and SDE thresholds before target-file prediction. The SDE thresholds are selected from (α,β) configurations in {0,1,2,3,4}, excluding (0,0), using local window statistics and RMSD-based calibration sweeps. Both wafer and FPGA calibration files support (α,β)=(2,2), which is then fixed for all remaining experiments. Experiments on industrial wafer data and actual FPGA silicon data, repeated across four random seeds, show that K-SDE reduces RMSD by 15.84% for wafer data and 13.03% for FPGA data relative to k-means sampling, while S-SDE reduces RMSD by 16.26% and 8.63% relative to stratified sampling.

Signals doi: 10.3390/signals7030052

Authors: AnuraagChandra Singh Thakur Masudul Imtiaz

Automatic modulation classification (AMC) is a core capability for spectrum monitoring, adaptive receivers, and electronic support. Most radio-frequency machine learning (RFML) studies train multi-class classifiers on benchmark datasets that contain a single modulation per recording at baseband. In operational settings, however, the objective is often to detect only a small set of signals of interest, making large multi-class models unnecessarily expensive to train and deploy. In addition, multi-class formulations can increase false-alarm risk due to confusion among non-essential classes and may allocate model capacity inefficiently to distinctions that are irrelevant for the operational objective. This paper investigates an alternative workflow based on targeted binary transformer detectors and evaluates their robustness under practical RF complications. Using the RadioML 2018.01A dataset, we construct binary detection tasks with BPSK as the signal of interest and introduce three increasingly realistic conditions: (i) center-frequency shifts away from baseband, (ii) sampling-rate mismatches via decimation and interpolation, and (iii) multi-signal mixtures where modulations co-occur either in frequency (simultaneous transmissions) or in time (temporal concatenation). The results show that baseband-trained detectors do not generalize to center-frequency-shifted signals, and multi-signal interference can cause complete detection failure unless explicitly modeled during training. We investigate early-exit transformer inference to reduce computation on high-confidence examples, showing it maintains (and occasionally improves) detection performance. We also evaluate inter-modulation transfer learning and intra-modulation adaptation from baseband to mixed- and multi-signal scenarios.

Signals doi: 10.3390/signals7030051

Authors: Yang Shen Xubin Liang Xiaolin Hu Shuping Wang

Similar to a web search engine, we have developed a computer-based acoustic search engine tailored for the critical scenario of high-speed aircraft ground explosion monitoring, addressing the long-standing challenge of real-time localization for such high-impact events. Unlike conventional acoustic source localization techniques, our method uniquely resolves the separation and localization of multiple aliasing events, which are prevalent in high-speed aircraft explosion scenarios due to complex shock wave propagation and overlapping signatures. We first calculate the waveforms of all possible acoustic sources over 2D grids. Then, a dimensionality reduction method and fast search technology are applied to the database. Once a high-speed aircraft ground explosion occurs, the real-time system returns detection feedback by matching real-time data with the pre-established search database. Different from other artificial intelligence (AI)-based approaches, the acoustic search engine can handle multiple aliased acoustic events in real time and does not require any prior information or input parameters—a key advantage for emergency response to high-speed aircraft explosions where predefined parameters are often unavailable. Both synthetic tests and field data applications (using actual acoustic records from high-speed aircraft ground explosion experiments) demonstrate the method’s credibility in detecting and localizing multiple acoustic sources.

Signals doi: 10.3390/signals7030050

Authors: Rylen Garlitz Allen Shamsi Ratree Wayland

Speech emotion recognition systems are typically trained on audio sampled at conventional bandwidths that exclude frequencies above approximately 8 kHz, yet the contribution of extended high-frequency information to vocal emotion recognition remains unclear. This study examines how spectral bandwidth influences automatic emotion classification using the RAVDESS corpus of acted speech and song. Recordings were low-pass filtered to simulate multiple bandwidth conditions (8, 12, and 16 kHz, along with the original full-bandwidth signal), and classification was performed using a Random Forest model trained on mel-spectral features. In addition to classification accuracy, we analyzed permutation-based spectral feature importance and the geometry of the classifier’s posterior-probability space. Bandwidth restriction had relatively modest effects on classification accuracy overall, with mean accuracy ranging from approximately 55% to 77% across conditions, although its impact was greater for speech than for song. Feature-importance analyses indicated that the model depends primarily on low- and mid-frequency spectral information, whereas higher-frequency and EHF regions show increased importance when available. Geometry analyses showed no reliable evidence that bandwidth altered the global structure of the stimulus-level emotion space, although spectral truncation reduced separability for certain emotion contrasts, particularly in speech at normal emotional intensity. These results indicate that most acoustic information supporting categorical emotion recognition resides in lower spectral regions, while EHF information provides supplementary acoustic information that may refine some emotional distinctions under specific conditions.

Signals doi: 10.3390/signals7030049

Authors: Ian G. Bearden Valentin Buchakchiev Daniel Ivanov Mira Gencheva Venelin Kozhuharov Yury A. Melikyan

Silicon photomultipliers (SiPMs) have become widely used as photodetectors in high-energy physics, nuclear physics, medical imaging, and space applications. In many of these fields, SiPMs are required to operate in high-radiation environments, which are notoriously problematic for silicon sensors. For this reason, it is essential to study the changes in their performance characteristics after exposure to radiation. In this study, a number of SiPM samples were exposed to non-uniform radiation at the CHARM facility at CERN. Half of the samples were operated above breakdown during the test, while others remained off. Intermittent measurements allowed for tracking the changes in I-V curves and signal shapes during the irradiation itself. The focus was on detecting differences in irradiation damage between the operational and non-operational SiPM samples. The I-V curves and signal shapes in both cases for three different types of SiPM are presented, and a comparison is made.

Signals doi: 10.3390/signals7030048

Authors: Huijie Zhu Wei Wang Cui Yang Youjun Xiang Jiawei Li Yuheng Xu

Frequency-hopping (FH) signals appear as small rectangular pulses in time-frequency spectrograms. At low signal-to-noise ratios (SNRs), noise along the frequency axis, caused by short-time Fourier transform (STFT) spectral leakage, blurs pulse boundaries, while the varying scales of hop rectangles exceed the capacity of a single receptive field. This paper presents YOLO11-FH, a modified YOLO11 detector that introduces two signal-processing-motivated modules. A FreqSmoothBlock (FSB) uses a (3,1) depthwise convolution to smooth exclusively along the frequency axis, while adding only 5C parameters. A TFMultiResBlock (TFMRB) fuses three parallel dilated convolution branches (dilation rates of 1, 2, and 3) to cover different hop scales, replacing a heavier C3k2 module. The detection head is further simplified by halving the Bottleneck repeat count and disabling the deep submodule at the P5 scale. On a simulated FH dataset (SNRs ranging from −15 dB to −10 dB, five jamming types), YOLO11-FH achieves 96.04% mean average precision (mAP)@0.5 and 76.18% mAP@0.5:0.95, outperforming the YOLO11n baseline by 0.95 and 2.91 percentage points (pp) with 2.9% fewer parameters.

Signals doi: 10.3390/signals7030047

Authors: Peiming Zhang Jiangang Lu Jiajia Fu Xinyue Di Kai Fang Jie Tang Cui Yang

Accurately capturing spatiotemporal dependencies and enabling effective decision support are core challenges in Intelligent Transportation Systems (ITS). Existing research often treats traffic prediction and path planning as isolated tasks. Moreover, mainstream prediction models struggle with long-term periodic patterns, while Reinforcement Learning (RL)-based planning often suffers from inefficient exploration in sparse topologies. To address these issues, this paper proposes a unified framework combining a spectral-temporal Graph Neural Network (GNN) and bi-directional decoding RL. Specifically, a time-frequency dual-stream adaptive learning module is introduced for prediction. Fast Fourier Transform (FFT) and Gated Recurrent Unit (GRU) are employed to capture global frequency periodicities and local temporal dynamics, respectively. Their adaptive fusion effectively mitigates the long-sequence information forgetting problem. For path planning, the task is formulated as sequence generation. A graph-aware attention encoder with adjacency masking is designed, and heuristic feature embeddings are incorporated to guide efficient exploration. Furthermore, a bi-directional autoregressive decoding strategy enhances robustness against topological bottlenecks. On PEMSD4 and PEMSD8, the proposed predictor achieves MAE/RMSE/MAPE values of 18.211/30.433/12.006 and 13.587/23.566/8.955, respectively. Path-planning simulations on the PEMSD4-derived sparse topology further demonstrate stable bi-directional RL optimization, faster convergence with heuristic guidance, and a sparsity-aware encoder that reduces redundant attention interactions in sparse road networks. These results validate the effectiveness of the proposed “predict-then-plan” paradigm.

Signals doi: 10.3390/signals7030046

Authors: Chia-Yen Yang Fan-Ning Kuo Hsin-Yung Chen

Despite being distinct clinical entities, major depressive disorder (MDD) and Parkinson’s disease (PD) have some shared physiological pathways, including mitochondrial dysfunction and inflammation. Our interest was whether these common physiological mechanisms are reflected in brain activity variations as well. Therefore, this study aimed to identify common characteristics in resting-state electroencephalography (EEG) between the conditions by comparing features among patients with MDD, PD, and healthy controls. The methodology comprised two stages: analyzing differences between patients and healthy individuals and exploring consistent trends between PD and MDD, based on EEG data from PRED + CT database. Age-corrected regression analysis of five EEG features revealed PD and MDD had the following overlapping features: shared abnormalities in theta, alpha and beta relative power, as well as sample entropy in the delta (centroparietal, temporal, and parietal areas), theta (parieto-occipital), and gamma (central) bands. Furthermore, interhemispheric asymmetry was evident across all bands, especially in the frontal and centroparietal regions. When combining these findings with their directional trends (positive or negative), common EEG features included increased theta and decreased alpha-beta power, along with increased parieto-occipital and reduced gamma entropy at FCz. These findings suggest shared EEG markers between PD and MDD, supporting the potential for efficient neurological disorder diagnosis.

Signals doi: 10.3390/signals7030045

Authors: Cristiano Fidani Alessandro Piscini Massimo Calcara Gianfranco Cianchini Maurizio Soldani Angelo De Santis Dario Sabbagh Martina Orlando Loredana Perrone

A thermal infrared (TIR) camera was installed at Mefite Lake in Valle d’Ansanto, Irpinia (Italy), to assess whether small variations in cold CO2 flux can be resolved thermally. To our knowledge, this is the first systematic attempt to extract short-period degassing dynamics from TIR data at Mefite. Infrared thermal images taken over a three-hour nighttime interval revealed the spatial distribution and extent of natural CO2 emissions. The high sampling frequency of one minute detected unexpected thermal variability from the source. The extent of temperature variations across the entire site reached almost 3 °C, with durations typically ranging from a few minutes to tens of minutes. Spectral analysis of the temperature time series reported a 1/f-type noise pattern, with significant periods of 2–3 min, 5 min, 26 min, and 61 min observed at different locations. Further intermediate periods were observed at individual points. Differences and delays in temperature variations appeared to be related to distance from the structure’s centre and the presence of water. These temperature fluctuations were interpreted as changes in the gaseous emission flow caused by a few kPa of CO2 escaping due to pressure variations. The gas thermally interacts with the underlying soil, adding or removing heat at the surface. These results demonstrate that high-frequency infrared thermography provides a sensitive and practical tool for quantifying short-term flux variability at natural CO2 vents and for improving the characterisation of their degassing dynamics.

Signals doi: 10.3390/signals7030044

Authors: Prajat Paul Mohamed Mehfoud Bouh Manan Vinod Shah Forhad Hossain Ashir Ahmed

Automatic speech recognition has advanced rapidly for high-resource languages, yet performance remains limited for low-resource languages such as Bangla, particularly in telehealth settings. Most systems rely on a standardized 16 kHz sampling rate, a design choice despite evidence that Bangla contains sibilant fricatives and other phonetic cues with substantial high-frequency energy that may be suppressed under bandwidth and latency constraints. This study evaluates audio sampling rate as a controllable signal-level parameter for Bangla telehealth ASR to identify an empirically grounded operating range balancing transcription accuracy, execution time, and network bandwidth. Twenty real-world Bangla doctor–patient consultations were deterministically resampled to 55 configurations between 8 kHz and 32 kHz and transcribed using a fixed cloud-based ASR system. Session-level Word Error Rate, execution latency, payload bandwidth, and high-frequency phonetic content were analyzed using a composite sibilant-likelihood score. WER decreased from 0.338 at 8 kHz to a local minimum of 0.232 at 18.75 kHz, with gains plateauing beyond this range despite substantial bandwidth increases. Elbow-point, Pareto frontier, weighted scoring, and Minimum Acceptable Trade-off analyses converged on an optimal region between 17.25 and 18.75 kHz, demonstrating that sampling rate optimization improves ASR accuracy without proportional resource costs in telehealth settings.

Signals doi: 10.3390/signals7030043

Authors: Sirine Aouine Chaieb Mahdi Abdelkarim Majdi Bahrouni Ali Gharsallah

This paper presents a compact, single-layer substrate-integrated waveguide (SIW) bandpass filter for 5.8 GHz WiMAX applications. The filter achieves an improved performance trade-off through a novel hybrid design strategy that combines central vertical perturbation vias with symmetrically etched complementary split-ring resonators (CSRRs). This configuration implements a hybrid magnetic–electric perturbation within a single cavity, enabling simultaneous control of electric and magnetic field confinement. The proposed topology achieves an optimized balance among unloaded quality factor Qu, insertion loss, selectivity, and structural simplicity. Through targeted intra-cavity field manipulation, the filter attains a Qu of 239.7, a narrow fractional bandwidth of 3.08% (5.75–5.93 GHz), and a low insertion loss of 1.12 dB. It also delivers enhanced selectivity compared to conventional single-cavity designs and performs competitively with multi-resonator architectures. An equivalent circuit model accurately captures the via–CSRR interaction and agrees closely with full-wave electromagnetic simulations. Experimental results confirm excellent return loss and robust performance across the entire WiMAX band (5.725–5.850 GHz). Thus, the proposed filter offers a practical, high-performance, and manufacturable solution for selective RF front-end applications.

Signals doi: 10.3390/signals7030042

Authors: Mostafa Moallim Taqwa A. Alhaj Fatin A. Elhaj Inshirah Idris Tasneem Darwish

Audio forgery has emerged as a significant security and forensic challenge, driven by rapid advances in generative artificial intelligence and the widespread availability of audio editing tools, which enable the creation of highly realistic manipulated speech with minimal technical expertise. Existing approaches predominantly operate at the file level, providing only coarse binary decisions without identifying when or where manipulation occurs. This study addresses fine-grained temporal localization through a unified frame-level localization framework. We introduce a controlled forgery generation framework derived from the TIMIT speech corpus, applying atomic, localized manipulations under strict temporal constraints and producing precise frame-level annotations across diverse manipulation types. Building on this dataset, we then propose a transform-agnostic localization-driven detection approach using temporal inconsistency modeling, enabling unified analysis across heterogeneous manipulations at frame-level resolution. To analyze forensic evidence, we present an evidence-stratified modeling paradigm comparing three complementary strategies: a handcrafted anomaly-based method, a deep localization model leveraging pretrained wav2vec 2.0 representations, and a hybrid approach combining both through confidence-aware fusion and temporal consistency reinforcement. A systematic experimental analysis evaluates the effects of representation adaptation, hybrid fusion, and manipulation type on detection and localization performance. Results show that handcrafted features are insufficient for reliable frame-level localization, while task-adapted wav2vec 2.0 achieves strong and consistent performance. The hybrid approach does not consistently improve frame-level accuracy but yields substantial gains in segment-level localization by enforcing temporal coherence. Per-transform analysis confirms robust performance across most manipulations, with deletion-based operations remaining the most challenging.

Signals doi: 10.3390/signals7030041

Authors: Joshua Turnipseed Benedito J. B. Fonseca

This paper investigates how well eGeMAPS features can be used to classify depression from a patient’s speech audio samples through the use of statistical resampling methods. We use permutation tests to evaluate, with high confidence, whether eGeMAPS features and the speaker’s depression status are dependent. We use bootstrap confidence intervals to test, with high confidence, whether eGeMAPS features are able to better discriminate depression in male speakers than in female speakers. Lastly, we compare the detection power of different subsets of the eGeMAPS features. We use an open-source dataset of depressed and non-depressed speakers (E-DAIC), an open-source audio feature extractor (eGeMAPS), and open-source machine learning classifiers (WEKA) to enable replication of results and establish a baseline for future studies.

Signals doi: 10.3390/signals7030040

Authors: Maged A. Aldhaeebi Abdulbaset Ali Thamer S. Almoneef

This paper presents a non-imaging landmine detection system that integrates a highly sensitive multiple-input multiple-output (MIMO) microwave sensor with a machine learning (ML) classifier for automated classification. The proposed sensor consists of two circular patch elements fed with two ports designed in a unique configuration, comprising a dual loop with a cross dipole, for enhancing sensitivity to changes in the environmental electrical properties (dielectric constant and electrical conductivity) induced by buried metallic objects. It operates in dual bands of 1.58 GHz and 1.75 GHz, within the operating frequency range of 1.3 to 2 GHz. The system’s performance was assessed using full-wave simulations and experimental measurements, involving a sand-filled foam container with a metal surrogate landmine placed at different depths. The sensor’s performance was evaluated by monitoring changes in the magnitude and phase of the reflection coefficient (S11) and the transmission coefficient (S21). The acquired scattering parameters data were processed using a Support Vector Machine (SVM) algorithm for automated classification. Results demonstrate the sensor’s high capability in detecting metallic targets at various depths and standoff distances. Compared to conventional imaging technologies, this system offers significant advantages in cost, simplicity, and ease of data processing. The SVM models trained on measurement data with proper feature selection showed a high level of agreement with their counterparts trained on simulation data. Stratified k-fold cross-validation was used to improve the reliability of accuracy metrics, with results showing 85% or higher mean accuracy in all classification scenarios.

Signals doi: 10.3390/signals7030039

Authors: Mladen Tomic Marko Gulic

Denoising one-dimensional signals by wavelet shrinkage critically depends on the choice of wavelet basis, yet basis selection is often guided by heuristics rather than explicit statistical criteria. This paper investigates the relationship between wavelet-basis properties and the shape of the probability density function (PDF) of the detail coefficients in the coarsest retained detail subband. On this basis, it proposes the shape of this PDF as a criterion for wavelet-basis selection. We hypothesize that, for a fixed decomposition depth, noise model, and shrinkage rule, a basis better matched to the signal’s local regularity produces a narrower and more sharply peaked coefficient PDF in this subband than a mismatched basis and can therefore serve as a data-driven indicator for basis selection. To evaluate the consistency of this proposal, we perform controlled hard-thresholding experiments on six canonical test signals, five wavelet bases, and additive white Gaussian noise. Although the test signals differ significantly in local regularity and features, the relationship between basis suitability and PDF shape is confirmed for each of them. Therefore, the results suggest that the proposed PDF-shape criterion is a valuable indicator for wavelet-basis selection.

Signals doi: 10.3390/signals7020038

Authors: Mahmoud Elomda Ahmed A. Ibrahim Mahmoud Abdelaziz

This paper presents a hardware-accelerated signal processing architecture for OFDM-based vehicular networks that integrates crypto-agile adaptive encryption on a Xilinx Kintex-7 FPGA. The encryption layer is tightly coupled to the OFDM modulation/demodulation pipeline, enabling secure real-time signal processing for V2X communications without disrupting the baseband chain. A context-aware pre-selection unit dynamically selects among hardware cipher primitives based on latency constraints, security requirements, and channel conditions. The current prototype implements and synthesizes AES-128 as the primary block cipher, while ASCON (NIST lightweight AEAD) and Keccak (SHA-3 foundation) are validated through RTL simulation and architectural integration, demonstrating crypto-agility across block, AEAD, and sponge-based primitives. DES is retained solely as a legacy reference for backward-compatibility evaluation and is not recommended for secure V2X deployment. The design adopts a modular decoupling strategy in which cryptographic engines interface with a unified buffering and interleaving subsystem, enabling hardware-based single-cycle cipher switching without partial reconfiguration. FPGA results demonstrate sub-microsecond cryptographic processing latencies with moderate resource utilization, preserving the timing budget of latency-sensitive vehicular services. AES-128 provides standard-strength encryption, while ASCON and Keccak offer lightweight and sponge-based alternatives suited to constrained IoV platforms. Specifically, the implemented AES-128 core achieves a throughput of 1.02 Gbps with a switching latency of 86 ns, verified across 10 randomized transitions with a 99.99% success rate and zero data corruption. The ASCON and Keccak cores attain throughput-to-area efficiencies of 2.01 and 1.47 Mbps/LUT, respectively, at a unified clock frequency of 50 MHz. All acronyms are defined at first use and a complete list of abbreviations is provided prior to the reference section.

Signals doi: 10.3390/signals7020037

Authors: Udara Jayasinghe Anil Fernando

Quantum image transmission over noisy communication channels remains a challenge due to the fragility of quantum states and their susceptibility to channel impairments. Existing quantum encoding schemes often exhibit limited noise resilience, while advanced approaches introduce computational and implementation complexity. To address these limitations, this paper proposes a circular polarization-based quantum encoding framework for image transmission over error-prone channels. In the proposed approach, source images are compressed and source-encoded using standard image coding formats, including the joint photographic experts group (JPEG) standard and the high-efficiency image file format (HEIF), and converted into classical bitstreams. The resulting bitstreams are protected using channel coding and mapped onto quantum states via circular polarization representations, where left- and right-hand circularly polarized states encode binary information. The encoded quantum states are transmitted over noisy quantum channels to model channel impairments. At the receiver, appropriate quantum decoding and channel decoding operations are applied to recover the classical bitstream, followed by source decoding to reconstruct the image. The performance of the proposed framework is evaluated using image quality metrics, including peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and universal quality index (UQI). Simulation results demonstrate that the proposed circular polarization-based encoding scheme outperforms existing quantum image encoding techniques, achieving channel SNR gains of 4 dB over state-of-the-art Hadamard-based encoding and 3 dB over frequency-domain quantum encoding methods under severe noise conditions. These results indicate that circular polarization-based quantum encoding provides improved noise robustness and reconstruction fidelity for practical quantum image transmission systems.

Signals doi: 10.3390/signals7020036

Authors: Domingos Sávio Tavares Mendes Junior Rafael Suzuki Bayma Alexandre Luiz Amarante Mesquita

This study investigated the influence of hyperparameter selection within an EEMD–AR–SVM framework for bearing fault diagnosis under constant- and variable-speed operating conditions. Two preprocessing configurations, namely, Method 1, in which EEMD was applied after segmentation, and Method 2, in which EEMD preceded segmentation, were evaluated under three rotational regimes—constant speed, acceleration (Test A), and deceleration (Test B)—while number of Intrinsic Mode Functions (N), autoregressive model order (L), and segment length were systematically varied towards identifying combinations that maximized classification accuracy. The results showed the methods achieved 100% accuracy under constant-speed operation. However, Method 2 consistently outperformed Method 1 under nonstationary regimes, reaching 94.12% accuracy during acceleration and 95.00% during deceleration. The outer race remained the most challenging fault type, although its separability substantially improved when EEMD was performed prior to segmentation. The findings demonstrated, in a clear and interpretable manner, that the empirical choice of N and L directly affects classifier accuracy in stationary and nonstationary scenarios and the order of preprocessing steps plays a decisive role in diagnostic reliability. Such contributions provide a reproducible methodological basis for advancing vibration-based fault diagnosis and support the development of interpretable, high-performance predictive maintenance strategies for industrial environments.

Signals doi: 10.3390/signals7020035

Authors: Yun Zhou Fulai Wang Nan Jiang Zhanling Wang Chen Pang Lei Zhang Yongzhen Li Ping Wang

Jamming pattern identification is a crucial prerequisite for countering jamming. Combined jamming exhibits complex structures and diverse forms, making it difficult for traditional identification methods to extract suitable and stable features for effective discrimination. To address this challenge, this paper proposes a combined jamming identification method based on joint modulation of linear frequency modulation, phase coding and phase coding frequency modulation (LFM-PC-PCFM) waveforms. Building upon the time–frequency entropy features of combined interference, this method enhances the separability of jamming features in the radar-transmitted waveform dimension. The experiment employed the SVM classification algorithm based on particle swarm optimization for validation. Experiments demonstrate that the combined jamming recognition method under LFM-PC-PCFM waveform modulation achieves higher and more stable recognition accuracy than traditional LFM single-waveform modulation under jamming-to-noise ratios ranging from −10 dB to 30 dB.

Signals doi: 10.3390/signals7020034

Authors: Tamoor Aziz Chalie Charoenlarpnopparut Srijidtra Mahapakulchai Babatunde Oluwaseun Ajayi Mayowa Emmanuel Bamisaye

Diabetic retinopathy is a prevalent condition and a major public health concern due to its detrimental impact on eyesight. Diabetes is a root cause of its development and damages small blood vessels caused by prolonged high blood sugar levels. The degenerative consequences of diabetic retinopathy are irrevocable if not diagnosed in the early stages of its progression. This ailment triggers the development of retinal lesions, which can be identified for diagnosis and prognosis. However, lesion detection is challenging due to their similarity in intensity profiles to other retinal features, inconsistent sizes, and random locations. This research evaluates a custom deep learning network for classifying retinal images and compares it with the state-of-the-art classifiers. The novel preprocessing method is introduced to reduce the complexity of the diagnostic process and to enhance classification performance by adaptively enhancing images. Despite being a shallow network, the proposed model yields competitive results with an accuracy of 87.66% and an F1-score of 0.78. The evaluation metrics indicate that class imbalance affects the performance of the proposed model despite using the weighted cross-entropy loss. The future contribution will be the inclusion of generative adversarial networks for generating synthetic images to balance the dataset. This research aims to develop a robust computer-aided diagnostic system as a second interpreter for ophthalmologists during the diagnosis and prognosis stages.

Signals doi: 10.3390/signals7020033

Authors: Jan-Ray Liao Wen Lin Li-Wen Chang

In recent years, image denoising has seen a shift from traditional non-local self-similarity methods like BM3D to deep-learning based approaches that use learnable convolutions and attention mechanisms. While pixel-level attention is effective at capturing long-range relationships similar to non-local self-similarity based methods, it incurs extremely high computational costs that scale quadratically with image resolution. As an alternative, channel-wise attention is resolution-independent and computationally efficient but may miss crucial spatial details. In this paper, an adjustable attention mechanism is introduced that bridges the gap between pixel and channel attentions. In the proposed model, average pooling and variable-size convolutions are added before attention calculation to adjust spatial resolution and, thus, allow dynamical adjustment of computational complexity. This adjustable attention is applied in a transformer-based U-Net architecture and achieves performance comparable to state-of-the-art methods in both real and Gaussian blind denoising tasks. To be more concrete, the proposed method achieves a Peak Signal-to-Noise Ratio of 39.65 dB and a Structural Similarity Index Measure of 0.913 on the Smartphone Image Denoising Dataset. Therefore, the proposed method demonstrates a balance between efficiency and denoising quality.

Signals doi: 10.3390/signals7020032

Authors: Xin-Qi Liao Yih-Min Chen

Compared to DVB-S2, DVB-S2X features a more intricate signaling structure. These signaling fields are employed not only in standard frames but are also frequently utilized within superframe structures. While rapid synchronization and decoding of these fields are critical, utilizing brute-force search methods incurs prohibitive computational costs. Therefore, this paper proposes a Joint Maximum Likelihood (JML) detection model tailored for the Fast Walsh–Hadamard Transform (FWHT). This approach allows for simultaneous synchronization and decoding while reducing number of real addition operations per codeword by approximately 15 times compared to brute-force methods. Consequently, the proposed architecture provides a highly efficient solution applicable to DVB-S2X and backward compatible with DVB-S2.

Signals doi: 10.3390/signals7020031

Authors: Lifang Fan Shaomin Liang Yanpeng Liu Guangbo Xiang Wei Zhang Xuexi Min

Airborne electromagnetic (AEM) surveys map subsurface electrical structures by deploying transmitter and receiver coils on an airborne platform. However, platform-induced vibrations are transmitted to the sensors, generating strong motion-induced noise that severely degrades signal quality. To mitigate such noise, this study proposed the use of granular materials as a cushioning medium. An impact model based on the Discrete Element Method (DEM) was developed and validated against drop-weight experiments. Both granular material properties and impactor characteristics were investigated. The study examined the cushioning effects on both the base plate and the impactor under impact loading, and the sensitivity of key parameters was evaluated. The results showed that granular properties had minimal influence on the impactor peak force. Increasing particle Young’s modulus, density, or friction coefficient led to higher peak forces on the base plate, with Young’s modulus and density having significantly stronger effects than friction coefficient. Additionally, both the impactor size and velocity correlate positively with the peak forces transmitted to the base plate and experienced by the impactor. Under thin layer conditions, the impactor force was more sensitive to impact parameters, while in thick layers it was mainly determined by particle rearrangement and energy dissipation mechanisms. These findings reveal the mechanisms governing granular cushioning and provide a theoretical basis for vibration isolation design in AEM systems to preserve high-fidelity signals.

Signals doi: 10.3390/signals7020030

Authors: Ziqiang Zhu Haigen Zhou Ao Wei Junfeng Yuan Handong Tan Manping Yang Zuoxi Jiang Marco Alfano

Airborne electromagnetic detection systems are highly susceptible to low-frequency motion-induced noise, which significantly degrades the extraction of weak geological signals. Conventional signal processing methods alone are often insufficient to suppress mechanically induced vibration noise, resulting in signal distortion and reduced detection reliability. To address this limitation, this study proposes an active noise suppression strategy that integrates mechanical vibration isolation with advanced signal processing. A pneumatic vibration isolation platform based on a cable-driven parallel robot (CDPR) architecture is developed to achieve precise orientation correction and effective vibration isolation. The system employs kinematic modeling and a servo-controlled pneumatic cylinder driven by a proportional directional valve to enable accurate dynamic regulation. Numerical simulations conducted in the Advanced Modeling and Simulation Environment (AMESim), combined with proportional–integral–derivative (PID) control, demonstrate that piston displacement overshoot is constrained within 0.2 mm. Furthermore, targeted filtering techniques are applied to enhance signal quality. Experimental results show that the response time for continuous step input is 0.18–0.2 s, with a steady-state error below 0.3 mm, confirming robust control performance. The proposed framework provides an effective low-noise solution for airborne electromagnetic detection and can improve survey reliability in deep resource exploration.

Signals doi: 10.3390/signals7020029

Authors: Dino Otero Ariel Amadio Leandro Robles Dávila Marcos Maillot Cristian Bonini Walter Legnani

The Virial theorem has been applied with considerable success in various fields of natural sciences. This work proposes an extension of the theorem applied to discrete data series. This application will be called the Virial theorem extension and can be applied to the numerical solution of nonlinear dynamic systems represented by difference equations, such as logistic, discubic and random number generators, the numerical solution of differential equations like the nonlinear double pendulum and a series of pseudorandom numbers and its reciprocals. For this purpose, a coefficient was derived from the discrete Virial formalism. This coefficient can be used to detect when a time series is obtained as the solution of a differential equation, in which case the coefficient is close to 1, and when the data come from other sources, in which case it takes different values. With reference to chaotic dynamic systems, the discrete Virial coefficient shows the feasibility in the detection of a change in behavior, as an alternative to the traditional calculation of Lyapunov exponents, and it is a thousand times faster. The convergence speed of the final value of the discrete Virial coefficient of a dynamic system in a non-chaotic regime is between one and five orders of magnitude greater than in the chaotic regime, thus extending results in non-Hamiltonian systems, previously found by another author in Hamiltonian systems. The results obtained show that the proposal characterizes and distinguishes different types of behavior from the series under study. It also shows great sensitivity to the evolution of the series, even anticipating critical points. The proposed method to construct the discrete Virial extension does not require the existence of a Hamiltonian, which allows its application to a series obtained experimentally or from any differential equation. From a general point of view, this research shows a series of properties that can be reinterpreted in light of the discrete Virial coefficient, providing a novel and versatile tool, given its minimal applicability requirements. For pseudorandom number series, the extension reveals a consistent, quasi-mirror behavior between its kinetic and potential factors, suggesting an underlying structural property.

Signals doi: 10.3390/signals7020028

Authors: Yutaka Yoshida Kiyoko Yokoyama

Sample-wise detection of P-, R-, and T-peaks in electrocardiograms (ECGs) is challenging because each peak type is sparsely represented (≈1:500 samples in a typical 10-s, 500-Hz ECG at 60 bpm), such that even a small number of false-positives (FPs) can markedly degrade positive predictive value (PPV) and limit the practicality of classifier-only approaches. This study proposes a lightweight ECG peak detection framework that combines binary classifiers with physiological temporal constraints (PTC) to address extreme sample-level class imbalance. Local morphological features are first evaluated using lightweight machine-learning models, among which XGBoost (XGB) exhibited the most stable score-ranking performance. Rather than directly thresholding classifier outputs, prediction scores are interpreted within the framework, which encodes physiological timing relationships. R-peaks are detected using score ranking combined with a refractory-period constraint, and the detected R-peaks serve as temporal landmarks for subsequent P- and T-peak detection within physiologically plausible time windows reflecting the P–QRS–T sequence. Quantitative evaluation was conducted using the Lobachevsky University Electrocardiography Database, hereafter referred to as LUDB. With a temporal tolerance of ±20 ms, the XGB-based system achieved an F1-score of 0.87 for R-peak detection (sensitivity 0.96, PPV 0.79), corresponding to approximately 9–10 true R-peaks with only 2–3 FP samples per 10-s segment. For P- and T-peaks, F1-scores of 0.70 and 0.69 were obtained, respectively. Additional evaluation on arrhythmic LUDB records demonstrated robust R-peak detection across rhythm types. In AF-related rhythms, where organized P waves are physiologically absent, the framework appropriately suppressed P-peak detections, with false-positive rates remaining below 0.31%. Qualitative application to ECG recordings from the PTB-XL database further demonstrated physiologically consistent behavior. These results indicate that reliable and interpretable ECG peak detection under extreme class imbalance can be achieved by integrating lightweight classifiers within the proposed framework, without reliance on complex deep learning architectures.

Signals doi: 10.3390/signals7020027

Authors: Ali Sarafnia M. Omair Ahmad M.N.S. Swamy

Distortion of speech in real-life communication is inevitable, affecting its quality. Conventionally, the effectiveness of a speech system in terms of the perceptual quality of the speech it produces has been assessed using a time-consuming subjective metric, the mean opinion score. There are a number of objective metrics that can be used instead of the mean opinion score to assess the perceptual quality of the speech signal. The objective of this paper is to propose and validate a new objective metric, the spectral entropy-based metric (SEM), designed to evaluate the perceptual quality of speech and perceptual naturalness by quantifying spectral coherence. While other metrics focus on intelligibility, this study aims to fill a gap in naturalness assessment. The core novelty of this work lies in offering a diagnostic perspective on spectral coherence, an indicator of speech naturalness that is often not explicitly addressed by other metrics. To demonstrate the effectiveness of the proposed metric in evaluating the perceptual quality, we consider fixed-beam and steerable-beam first-order differential microphone arrays. Compared with other objective metrics, it is shown that the proposed SEM is more sensitive to spectral coherence, a predominant indicator of the naturalness of the output speech signal of a speech system.

Signals doi: 10.3390/signals7020026

Authors: Md. Selim Al Mamun Fatema Akhter

The accurate positioning of location in indoor environment has become crucial in many location-based services, mainly where global positioning systems (GPSs) are unavailable or fail to navigate correctly. Conventional fingerprint-based approaches face challenges with instability, low accuracy, and being sensitive to changes in the environment. This study proposes a robust fingerprint-based machine learning (ML) model for dynamic environment indoor navigation in real time. The proposed model uses link quality indicator (LQI) values from IEEE 802.15.4 as fingerprints and supervised learning algorithms, showing high accuracy and a strong ability to adapt to changes in the environment. A room within a building floor has been regarded as the unit of location identification instead of the user’s exact coordinates to make the suggested model more relevant under practical conditions. The model was trained and tested using a real LQI dataset collected from varied indoor conditions to ensure the system can adapt effectively and operate consistently in dynamic environments and signal conditions. The results show that the proposed model surpasses fingerprinting indoor navigation in room detection accuracy and flexibility to environmental changes. An implemented prototype proved the real-time capability of the proposal in smart buildings, hospitals, and industrial IoT settings.

Signals doi: 10.3390/signals7020025

Authors: Peter Swartz Saiyu Ren Shuxia Sun James Martin

The Nyquist Folding Receiver is an architecture that uses Compressed Sensing to convert analog radio frequency signals into digital signals. Analog-to-Digital Converter architectures that implement Compressed Sensing are collectively known as Analog-to-Information. Sparse bandlimited analog signals with frequency bands above the Nyquist frequency of a traditional Analog-to-Digital Converter can be recovered by Analog-to-Information receivers. Recovery of these signals is affected by the selection of a Compressed Sensing recovery algorithm. Typical recovery algorithms selected for recovery of Nyquist Folding Receiver-compressed outputs use iterative methods to find the solution. This work presents a machine learning approach to signal reconstruction. The proposed method uses a neural network to learn the mapping from compressed samples to the original signal. The neural network is trained on a set of synthetic signals generated by a new open-source Analog-to-Information simulator called SpectraMelt. The results show that the neural network can effectively reconstruct the original signal from the compressed samples, achieving better performance than traditional iterative methods.

Signals doi: 10.3390/signals7020024

Authors: Huu-Son Nguyen Masaki Kurosawa Koichiro Ishibashi Ryou Tanaka Cong-Kha Pham Guanghao Sun

This study presents a harmonic-based method for non-contact heart rate (HR) estimation from continuous-wave (CW) Doppler radar signals, validated across multiple species including humans and small animals (cat). Traditional frequency-domain methods struggle when the HR fundamental frequency is weak or overlaps with respiratory components. The proposed approach addresses this by identifying three higher-order HR harmonics (2nd, 3rd, and 4th) then reconstructing the HR fundamental frequency from their integer ratios (3/2, 4/3, 2/1). The algorithm processes 20-s sliding windows (1-s overlap) using bandpass filtering to remove respiratory components and HR fundamental while preserving higher harmonics, followed by Power Spectral Density (PSD) analysis. When a complete harmonic set cannot be found, the proposed algorithm switches to harmonic pair detection, enhancing robustness when one harmonic is absent or attenuated. Besides, an adaptive tolerance mechanism enables detection under non-ideal conditions. The method was validated using a public human dataset and an experimental cat dataset with varied positions (supine/prone) and anesthesia levels (1–3% isoflurane). For humans, the algorithm achieved HR Accuracy consistently above 98% with an average RMSE of 1.33 bpm (MAPE: 1.29%, MAE: 0.86 bpm) and Bland-Altman bias below 0.9 bpm. For the cat dataset, performance was even better with HR Accuracy remaining above 99%, an average RMSE of 0.39 bpm (MAPE: 0.22%, MAE: 0.30 bpm), and bias below 0.14 bpm.

Signals doi: 10.3390/signals7020023

Authors: Longqing Li Wenjun Xie Deming Hu Jingke Nie Fei Xie Zhiping Huang Yongjie Zhao

Signal-to-noise ratio (SNR) estimation plays a crucial role in communication systems, directly impacting the quality and reliability of signal transmission. This paper proposes a novel deep learning framework aimed at enhancing the accuracy and robustness of SNR estimation. The framework converts received signals into time–frequency matrices as feature inputs, effectively capturing both temporal and spectral characteristics through time–frequency analysis. Extensive experimental results across an SNR range of −5 dB to 15 dB demonstrate that our method achieves a mean squared error (MSE) that closely approaches the theoretical Cramér–Rao bound (CRB), comparable to data-aided (DA) maximum likelihood methods. A quantitative analysis reveals that, even under challenging conditions, such as a low SNR of −5 dB, the model maintains superior accuracy with a mean absolute error (MAE) as low as 0.352, significantly outperforming traditional M2M4 and NDA estimators. The model’s performance was systematically evaluated in a wide range of scenarios, encompassing various signal modulation formats, upsampling factors, multipath fading channels, frequency offsets, phase shifts, and roll-off factors. The evaluation highlights its exceptional generalization capability and robustness, with high performance and stability maintained even in challenging and dynamic environments.

Signals doi: 10.3390/signals7020021

Authors: Yuan Xing Xuedong Ding Haowen Pan

Modern manufacturing quality control demands intelligent, adaptive inspection systems capable of real-time contamination detection with minimal computational overhead. We present a five-agent vision framework for material-aware contamination detection in manufacturing environments. The system comprises: a Material Classification Agent that identifies contamination type (fiber, sand, or mixed), three Material-Specific Detection Agents, each employing patch-based CNNs optimized for their respective material with dynamic patch size selection (128 px, 256 px, 384 px), and an Adaptation Agent that monitors performance and eliminates consistently failing patch size configurations. This hierarchical architecture enables intelligent routing to specialized detectors and continuous refinement through performance-driven adaptation. The Material Classification Agent achieves 98% accuracy in contamination type identification. Material-specific agents demonstrate F1-scores of 0.968 (fiber), 0.977 (sand), and 0.977 (mixed) with real-time inference (2.40–11.11 ms per 512 × 512 image). The Adaptation Agent implements selective patch size elimination: configurations failing quality thresholds (F1 < 0.5) across multiple evaluation cycles are removed from the detection pipeline. On the synthetic test split used in this study, comparative evaluation against PatchCore, WinCLIP, and PaDiM shows 3–45× higher F1-scores with superior accuracy–latency trade-offs, validating the efficacy of specialized material-aware architectures for manufacturing contamination detection.

Signals doi: 10.3390/signals7020022

Authors: Mawardi Bahri Marni Rezki Samsul Ariffin Abdul Karim Nasrullah Bachtiar Muhammad Zakir Bannu Addul Samad

The offset fractional Fourier transform (OFrFT) has emerged as a mathematical tool for time-frequency analysis of non-stationary signals. However, there has been relatively little research on investigating properties including the convolution theorem and the sampling formulas. This paper investigates the new form of convolution theorem and its product theorem associated with the OFrFT. We also establish a sampling formula related to the transformation. Finally, a simple example is displayed to verify the results related to the sampling formula for the OFrFT.

Signals doi: 10.3390/signals7020020

Authors: Yutaka Yoshida Kiyoko Yokoyama

Electrocardiography (ECG) is a key biosensing technique for assessing cardiac function and autonomic activity. Accurate detection of R-peaks and precise calculation of R-R intervals (RRIs) are essential for heart rate variability (HRV) analysis; however, automated detection algorithms remain vulnerable to local misdetections, such as false positives or missed beats (false negatives), caused by noise, baseline fluctuations, or waveform variability. Conventional correction approaches based on filter or threshold adjustment may introduce new errors outside the target region, highlighting the need for an intuitive and localized manual correction capability. To address this issue, we developed a prototype graphical user interface (GUI)-based ECG viewer implemented in Fortran for high computational efficiency. The system enables interactive insertion and deletion of detected R-peaks, with recalculation of the RRI time series and automatic updating of related analyses, including power spectral density, histograms, Lorenz plots, and polar plots. Validation using synthetic ECG signals at four sampling frequencies (125–1000 Hz) and three display time scales (2, 5, and 10 s) demonstrated correction errors below 0.7% and stable update times within 20–30 ms. When applied to real ECG recordings from the MIT-BIH Arrhythmia Database (records 115, 122, and 209; MLII lead), the GUI-derived RRIs achieved accuracies exceeding 0.985 at a strict ±10 ms tolerance and reached 1.000 at ±20 ms or higher, including recordings with frequent atrial premature contractions. These results indicate that the proposed system provides reliable feedback for localized correction of R-peak misdetections without altering the underlying ECG signal. The proposed algorithm may support future research and experimental applications in biosignal processing.

Signals doi: 10.3390/signals7020019

Authors: Alexey García Padilla Ivonne Bazán Trujillo Carlos A. Negreira Casares

Displacement estimation methods in strain elastography use ultrasound signals to estimate displacements and strain in soft tissue. Although several methods exist, systematic comparisons under controlled simulation conditions for lesions smaller than 5 mm are limited. Evaluating axial accuracy for superficial and intermediate depths in small breast cancer lesions is clinically important, as early-stage detection with existing techniques remains challenging. In this study, speckle tracking, the Doppler method, and the combined autocorrelation method (CAM) were used to estimate axial displacements from simulated elastography signals. The performance of these methods was assessed using the root mean square error (RMSE) for displacement field, strain, and tumor size estimation in a three-layer model comprising healthy tissue–tumor–healthy tissue. An extended analysis considering anatomically realistic tissue and motion artifacts conditions for the case of smallest lesion is presented. Finally, the CAM method, which obtained the best results, was assessed varying SNR/strain values. Simulation results show that CAM outperforms the other methods in displacement estimation across early-stage tumor sizes at both superficial and intermediate depths in all performed tests.

Signals doi: 10.3390/signals7010018

Authors: Malaz Alfa Fars Samann Thomas Schanze

Background: Electrocardiography (ECG), particularly the 12-lead configuration, is a crucial method for identifying heart rhythm abnormalities. However, its effectiveness can be reduced by noise contamination. State-of-the-art denoising methods based on neural networks have demonstrated promising performance in denoising complex biosignals like ECG. However, most of these methods have focused on denoising single-lead ECG recordings. Methods: This research aims to leverage the inherent correlation among multi-lead ECG signals. Therefore, a multi-lead convolutional denoising autoencoder (ML-CDAE) model is proposed, to learn more effective representations, leading simultaneously to improved denoising performance and enhanced quality of 12-lead ECG recordings. Results: The findings indicate that ML-CDAE consistently outperforms a single-lead convolutional denoising autoencoder (SL-CDAE) and fully convolutional denoising autoencoder (FCN-DAE) model in denoising ECG signals corrupted by a mixture of physical noises. In particular, the mean squared error (MSE) and signal-to-noise ratio improvement (SNRimp) are used as evaluation metrics to assess the performance. Conclusions: The strong correlation among multi-lead ECG signals can be leveraged not only to enhance the denoising performance of the ML-CDAE model but also to simultaneously denoise 12-lead ECG signals more successfully compared to both the SL-CDAE and FCN-DAE models.

Signals doi: 10.3390/signals7010017

Authors: Christos Kalogeropoulos Konstantinos Theofilatos Seferina Mavroudi

Electroencephalography (EEG) has transitioned from a subjective observational method into a data-intensive analytical field that utilises sophisticated algorithms and mathematical models. This review provides a holistic foundation by detailing the neurophysiological basis, recording techniques, and applications of EEG before providing a rigorous examination of traditional and modern analytical pillars. Statistical and Time-Series Analysis, Spectral and Time-Frequency Analysis, Spatial Analysis and Source Modelling, Connectivity and Network Analysis, and Nonlinear and Chaotic Analysis are explored. Afterwards, while acknowledging the historical role of Machine Learning (ML) and Deep Learning (DL) architectures, such as Support Vector Machines (SVMs) and Convolutional Neural Networks (CNNs), this review shifts the primary focus toward current state-of-the-art Artificial Intelligence (AI) trends. We place emphasis on the emergence of Foundation Models, including Large Language Models (LLMs) and Large Vision Models (LVMs), adapted for high-dimensional neural sequences. Finally, we explore the integration of Generative AI for data augmentation and review Explainable AI (XAI) frameworks designed to bridge the gap between “black-box” decoding and clinical interpretability. We conclude that the next generation of EEG analysis will likely converge into Neuro-Symbolic architectures, synergising the massive generative power of foundation models with the rigorous, rule-based interpretability of classical signal theory.

Signals doi: 10.3390/signals7010016

Authors: Manal Hilali Abdellah Ezzati Said Ben Alla Ahmed El Badaoui

Automatic emotion recognition based on EEG has been a key research frontier in recent years, involving the direct extraction of emotional states from brain dynamics. However, existing deep learning approaches often treat EEG either as a sequence or as a static spatial map, thereby failing to jointly capture the temporal evolution and spatial dependencies underlying emotional responses. To address this limitation, we propose an Interpretable Residual Spatio-Temporal Graph Attention Network (IRSTGANet) that integrates temporal convolutional encoding with residual graph-attention blocks. The temporal module enhances short-term EEG dynamics, while the graph-attention layers learn adaptive node connectivity relationships and preserve contextual information through residual links. Evaluated on the DEAP and SEED datasets, the proposed model achieved exceptional performance on valence and arousal, as well as four-class and nine-class classification on the DEAP dataset and on the three-class SEED dataset, exceeding state-of-the-art methods. These results demonstrate that combining temporal enhancement with residual graph attention yields both improved recognition performance and interpretable insights into emotion-related neural connectivity.

Signals doi: 10.3390/signals7010015

Authors: Seyed Reza Razavi Ashish H. Shah Zahra Moussavi

Despite successful revascularization, patients with non-ST elevation myocardial infarction (NSTEMI) remain at higher risk of mortality and morbidity. Accurately predicting mortality risk in this cohort can improve outcomes through timely interventions. This study for the first time predicts 1-year all-cause mortality in an NSTEMI cohort using features extracted primarily from the aortic pressure (AP) signal recorded during cardiac catheterization. We analyzed data from 497 NSTEMI patients (66.3 ± 12.9 years, 187 (37.6%) females) retrospectively. We developed three survival models, the multivariate Cox proportional hazards, DeepSurv, and random survival forest, to predict mortality. Then, used Shapley additive explanations (SHAP) to interpret the decision-making process of the best survival model. Using 5-fold stratified cross-validation, DeepSurv achieved an average C-index of 0.935, an IBS of 0.028, and a mean time-dependent AUC of 0.939, outperforming the other models. Ejection systolic time, ejection systolic period, the difference between systolic blood pressure and dicrotic notch pressure (DesP), skewness, the age-modified shock index, and myocardial oxygen supply/demand ratio were identified by SHAP as the most characteristic AP features. In conclusion, AP signal features offer valuable prognostic insight for predicting 1-year all-cause mortality in the NSTEMI population, leading to enhanced risk stratification and clinical decision-making.

Signals doi: 10.3390/signals7010014

Authors: Oussama Boufares Wajdi Saadaoui Mohamed Boussif

Foreground segmentation and background subtraction are critical components in many computer vision applications, such as intelligent video surveillance, urban security systems, and obstacle detection for autonomous vehicles. Although extensively studied over the past decades, these tasks remain challenging, particularly due to rapid illumination changes, dynamic backgrounds, cast shadows, and camera movements. The emergence of supervised deep learning-based methods has significantly enhanced performance, surpassing traditional approaches on the benchmark dataset CDnet2014. In this context, this paper provides a comprehensive review of recent supervised deep learning techniques applied to background subtraction, along with an in-depth comparative analysis of state-of-the-art approaches available on the official CDnet2014 results platform. Specifically, we examine several key architecture families, including convolutional neural networks (CNN and FCN), encoder–decoder models such as FgSegNet and Motion U-Net, adversarial frameworks (GAN), Transformer-based architectures, and hybrid methods combining intermittent semantic segmentation with rapid detection algorithms such as RT-SBS-v2. Beyond summarizing existing works, this review contributes a structured cross-family comparison under a unified benchmark, a focused analysis of performance behavior across challenging CDnet2014 scenarios, and a critical discussion of the trade-offs between segmentation accuracy, robustness, and computational efficiency for practical deployment.

Signals doi: 10.3390/signals7010013

Authors: Hussam Rostum József Vásárhelyi

Feature extraction and description are fundamental components of visual perception systems used in applications such as visual odometry, Simultaneous Localization and Mapping (SLAM), and autonomous navigation. In resource-constrained platforms, such as Unmanned Aerial Vehicles (UAVs), achieving real-time hardware acceleration on Field-Programmable Gate Arrays (FPGAs) is challenging. This work demonstrates an FPGA-based implementation of an adaptive ORB (Oriented FAST and Rotated BRIEF) feature extraction pipeline designed for high-throughput and energy-efficient embedded vision. The proposed architecture is a completely new design for the main algorithmic blocks of ORB, including the FAST (Features from Accelerated Segment Test) feature detector, Gaussian image filtering, moment computation, and descriptor generation. Adaptive mechanisms are introduced to dynamically adjust thresholds and filtering behavior, improving robustness under varying illumination conditions. The design is developed using a High-Level Synthesis (HLS) approach, where all processing modules are implemented as reusable hardware IP cores and integrated at the system level. The architecture is deployed and evaluated on two FPGA platforms, PYNQ-Z2 and KRIA KR260, and its performance is compared against CPU and GPU implementations using a dedicated C++ testbench based on OpenCV. Experimental results demonstrate significant improvements in throughput and energy efficiency while maintaining stable and scalable performance, making the proposed solution suitable for real-time embedded vision applications on UAVs and similar platforms. Notably, the FPGA implementation increases DSP utilization from 11% to 29% compared to the previous designs implemented by other researchers, effectively offloading computational tasks from general purpose logic (LUTs and FFs), reducing LUT usage by 6% and FF usage by 13%, while maintaining overall design stability, scalability, and acceptable thermal margins at 2.387 W. This work establishes a robust foundation for integrating the optimized ORB pipeline into larger drone systems and opens the door for future system-level enhancements.

Signals doi: 10.3390/signals7010012

Authors: Tomoki Matsunaga Hiroaki Saito

Multitrack music transcription is the task of converting music recordings into symbolic music representations that are assigned to individual instruments. This task requires simultaneous transcription of note onset and offset events for individual instruments. In addition, the limited resources of many transcription datasets make multitrack music transcription challenging. Thus, even state-of-the-art transcription systems are inadequate for applications requiring high accuracy. In this paper, we propose a framework to jointly transcribe onsets and frames for multiple instruments by integrating a deep learning architecture based on U-Net with an architecture based on Perceiver, which is a variant of the Transformer architecture. The proposed framework effectively detects the pitches of different instruments by employing the multi-layer combined frequency and periodicity (ML-CFP) with multilayered frequency-domain and quefrency-domain features as the input data representation. Our experiments demonstrate that the proposed multitrack music transcription system outperforms existing systems on five transcription datasets, including low-resource datasets. Furthermore, we evaluate the proposed system in terms of instrument type and show that the system provides high-quality transcription results for the predominant instruments.

Signals doi: 10.3390/signals7010011

Authors: Nefeli Chatzisavvidou Stefanos Papasotiriou Ioanna Vrachni Konstantinos Kantelis Petros Nicopolitidis Georgios Papadimitriou

With rapid advances in nanotechnology and synthetic biology, biological nanonetworks are emerging for biomedical and environmental applications within the Internet of Bio-NanoThings. While they rely on molecular communication, experimental validation remains limited, especially for non-ideal effects such as molecular accumulation. In this work, we present a novel table-top experimental system that emulates the core functionalities of a biological nanonetwork and is straightforward to reproduce in standard laboratory environments, also making it suitable for educational demonstrations. To the best of our knowledge, this is the first experimental platform that incorporates two end nodes capable of acting interchangeably as transmitter and receiver, thereby enabling true bidirectional molecular communication. Information transfer is realized through controlled release, advection and diffusion of molecules, using molecular concentration coding analogous to concentration shift keying, while the receiver decodes messages by comparing measured concentrations against predefined thresholds. Based on the measurements reported herein, the drop-based algorithm substantially outperforms the threshold-based scheme. Specifically, it reduces first-message latency by more than 2.5× across the tested volumes and reduces latest-message latency by up to 71%, providing approximately 3.7× better message delivery. A key experimental outcome is the observation of channel saturation: beyond a certain operating period, residual molecules accumulate and effectively saturate the medium, inhibiting reliable further message exchange until sufficient clearance occurs. This saturation-induced “channel memory” emerges as a fundamental practical constraint on sustained communication and achievable data rates. Overall, the proposed platform provides a scalable, controllable, and experimentally accessible testbed for systematically studying signal degradation, saturation, clearance dynamics, and throughput limits, thereby bridging the gap between theoretical models and practical implementations in the Internet of Bio-NanoThings era.

Signals doi: 10.3390/signals7010010

Authors: Yifan Liu Yonggang Xu Yanping Zhu Xue Zou Huaming Zhang

Bearing fault diagnosis is critical in rotating machinery, and collecting and analyzing vibration signals from faulty bearings is a widely employed method in fault diagnosis. To efficiently extract the information of periodic pulse from complex signals and accurately identify fault characteristic frequencies, this paper proposes a BSNA (Bispectral Slice Negentropy Analysis) method. This method leverages the nonlinear characteristics of bispectral analysis and the sensitivity of negentropy measures to transform one-dimensional signals into two-dimensional spectra. By utilizing the demodulation capability of the time-frequency modulation bispectrum, it highlights the relationship between resonance bands and modulation frequency, while maximizing the preservation of critical fault information and minimizing the impact of interference signals. The fault information contained in the slices is subsequently quantified using the CSNE (correlation spectral negentropy), which effectively captures the magnitude of periodic pulse energy. By calculating the CSNE of each modulation frequency slice and visualizing it, the energy distribution of periodic pulses within each slice can be effectively observed. The feasibility of this method in rolling bearing fault diagnosis has been validated through simulation analysis and experimental comparison. This approach enables the accurate identification of fault characteristic frequency and its harmonics, thereby significantly enhancing the accuracy and robustness of fault diagnosis, particularly in complex and noisy background environments.

Signals doi: 10.3390/signals7010009

Authors: Shaoxun Lu Huabo Liu

This paper focuses on the design of robust fusion estimators for multi-sensor systems experiencing constrained communications, model uncertainties, and random packet dropouts. To mitigate the impact of modeling errors, a sensitivity-penalized robust state estimator is employed at each local estimator. At the local fusion estimators, a centralized robust fusion estimation algorithm is derived by improving the cost function of the sensitivity-penalized estimator. The implementation of an event-triggered strategy effectively alleviates the burden on the communication channels linking the sensors and the fusion center. Moreover, the fusion estimator is capable of handling packet drops caused by unreliable communication channels, and the pseudo cross-covariance matrix is accordingly formulated. Sufficient conditions are derived to ensure the uniform boundedness of the estimation error for the proposed robust fusion estimator. Finally, simulation experiments using a tractor-car system validate the performance and advantages of the presented algorithm.

Signals doi: 10.3390/signals7010008

Authors: Faroq Awin Yasser Alginahi Esam Abdel-Raheem

This survey delves into the Firebug Swarm Optimization (FSO) algorithm, an advanced global optimization algorithm that plays a pivotal role in modern swarm intelligence optimization techniques. It explores the core principles of the FSO algorithm and examines the various hybrid variants developed to address complex optimization challenges. This survey also traces the evolution of swarm optimization methods, shedding light onto the natural phenomena and biological processes that have inspired these algorithms. Furthermore, it highlights the diverse real-world applications of the FSO algorithm, showcasing its effectiveness in fields such as engineering, data science, and artificial intelligence. To provide a comprehensive comparison, the survey includes a case study that evaluates the FSO algorithm’s performance against other existing algorithms. Lastly, the survey identifies key open research questions and suggests potential future directions for advancing the FSO algorithm and other nature-inspired optimization techniques, aiming to overcome current limitations and unlock new possibilities.

Signals doi: 10.3390/signals7010007

Authors: Ajan Ahmed Masudul H. Imtiaz

Self-supervised learning (SSL) models have achieved remarkable success in speaker verification tasks, yet their robustness to real-world audio degradation remains insufficiently characterized. This study presents a comprehensive analysis of how audio quality degradation affects three prominent SSL-based speaker verification systems (WavLM, Wav2Vec2, and HuBERT) across three diverse datasets: TIMIT, CHiME-6, and Common Voice. We systematically applied 21 degradation conditions spanning noise contamination (SNR levels from 0 to 20 dB), reverberation (RT60 from 0.3 to 1.0 s), and codec compression (various bit rates), then measured both objective audio quality metrics (PESQ, STOI, SNR, SegSNR, fwSNRseg, jitter, shimmer, HNR) and speaker verification performance metrics (EER, AUC-ROC, d-prime, minDCF). At the condition level, multiple regression with all eight quality metrics explained up to 80% of the variance in minDCF for HuBERT and 78% for WavLM, but only 35% for Wav2Vec2; EER predictability was lower (69%, 67%, and 28%, respectively). PESQ was the strongest single predictor for WavLM and HuBERT, while Shimmer showed the highest single-metric correlation for Wav2Vec2; fwSNRseg yielded the top single-metric R2 for WavLM, and PESQ for HuBERT and Wav2Vec2 (with much smaller gains for Wav2Vec2). WavLM and HuBERT exhibited more predictable quality-performance relationships compared to Wav2Vec2. These findings establish quantitative relationships between measurable audio quality and speaker verification accuracy at the condition level, though substantial within-condition variability limits utterance-level prediction accuracy.

Signals doi: 10.3390/signals7010006

Authors: Roman N. Ipanov Aleksey A. Komarov

To obtain radar images of a group of small space objects or to resolve individual elements of complex space objects in near-Earth orbit, a radar system must have high spatial resolution. High range resolution is achieved by using complex probing signals with a wide spectrum bandwidth. Achieving high angular resolution for small or complex space objects is based on the inverse synthetic aperture antenna effect. Among the various classes of complex signals, only two have found practical application in Inverse Synthetic Aperture Radar (ISAR) systems so far: the Linear Frequency-Modulated signal (chirp) and the Stepped-Frequency signal. Over the coherent integration interval of the echo signals, which corresponds to the ISAR aperture synthesis time, the combined correlation characteristics of the signal ensemble are analyzed. A high level of integral correlation noise in the ensemble of probing signals degrades the quality of the radar image. Therefore, a probing signal with a Zero Autocorrelation Zone (ZACZ) is highly relevant for ISAR applications. In this work, through simulation, radar images of a complex space object were obtained using both chirp and ZACZ probing signals. A comparative analysis of the correlation characteristics of the echo signals and the resulting radar images of the complex space object was performed.

Signals doi: 10.3390/signals7010005

Authors: Arunee Promsri

Leg dominance has been linked to an increased risk of lower-limb injuries in sports. This study examined bilateral asymmetry in muscle synergy patterns during one-leg stance on stable and multiaxial unstable surfaces. Twenty-five active young adults (25.6 ± 3.9 years) performed unipedal stance tasks on their dominant and non-dominant legs while surface electromyography (EMG) was recorded from seven lower-limb muscles per leg. Muscle synergies were extracted using non-negative matrix factorization (NMF), and structural similarity was assessed via cosine similarity with the Hungarian matching algorithm. Four consistent synergies were identified under both surface conditions, accounting for 88% of the total variance. On the stable surface, significant asymmetry in muscle weightings was observed in the rectus femoris (p = 0.030) for Synergy 1 and in the rectus femoris (p = 0.042), tibialis anterior (p = 0.024), peroneus longus (p = 0.023), and soleus (p = 0.006) for Synergy 2. On the unstable surface, asymmetry was evident in the biceps femoris (p = 0.048) for Synergy 2 and the rectus femoris (p = 0.045) for Synergy 3. Overall, dominance-related asymmetry was more pronounced under stable conditions and became more subtle as postural demand increased, revealing bilateral asymmetry in neuromuscular coordination during unipedal stance.

Signals doi: 10.3390/signals7010004

Authors: Innocent Tujyinama Bessam Abdulrazak Rachid Hedjam

Background and Objective: The accurate detection of sleep stages and disorders in older adults is essential for the effective diagnosis and treatment of sleep disorders affecting millions worldwide. Although Polysomnography (PSG) remains the primary method for monitoring sleep in medical settings, it is costly and time-consuming. Recent automated models have not fully explored and effectively fused the sleep features that are essential to identify sleep stages and disorders. This study proposes a novel automated model for detecting sleep stages and disorders in older adults by analyzing PSG recordings. PSG data include multiple channels, and the use of our proposed advanced methods reveals the potential correlations and complementary features across EEG, EOG, and EMG signals. Methods: In this study, we employed three novel advanced architectures, (1) CNNs, (2) CNNs with Bi-LSTM, and (3) CNNs with a transformer encoder, for the automatic classification of sleep stages and disorders using multichannel PSG data. The CNN extracts local features from RGB spectrogram images of EEG, EOG, and EMG signals individually, followed by an appropriate column-wise feature fusion block. The Bi-LSTM and transformer encoder are then used to learn and capture intra-epoch feature transition rules and dependencies. A residual connection is also applied to preserve the characteristics of the original joint feature maps and prevent gradient vanishing. Results: The experimental results in the CAP sleep database demonstrated that our proposed CNN with transformer encoder method outperformed standalone CNN, CNN with Bi-LSTM, and other advanced state-of-the-art methods in sleep stages and disorders classification. It achieves an accuracy of 95.2%, Cohen’s kappa of 93.6%, MF1 of 91.3%, and MGm of 95% for sleep staging, and an accuracy of 99.3%, Cohen’s kappa of 99.1%, MF1 of 99.2%, and MGm of 99.6% for disorder detection. Our model also achieves superior performance to other state-of-the-art approaches in the classification of N1, a stage known for its classification difficulty. Conclusions: To the best of our knowledge, we are the first group going beyond the standard to investigate and innovate a model architecture which is accurate and robust for classifying sleep stages and disorders in the elderly for both patient and non-patient subjects. Given its high performance, our method has the potential to be integrated and deployed into clinical routine care settings.

Signals doi: 10.3390/signals7010003

Authors: Alessandro Filisetti Pietro Bia Germana Luciani Margherita Losardo Riccardo Ardoino Antonio Manna

The absence of a recognizable P wave in an electrocardiogram (ECG) is a critical indicator for the diagnosis of atrial fibrillation (AF). An algorithm capable of distinguishing between physiological and pathological states in a short period of time could serve as a valuable tool for timely and effective diagnosis, even in a home setting. To achieve this goal, a deterministic algorithm is proposed. The Fantasia Database and the AF Termination Challenge Database were used for training the model. Subsequently, for the test session, a one-minute recording was extracted from the Autonomic Aging Dataset and the Long-Term AF Database. After band-pass filtering, characteristic points such as R-peaks and P waves were extracted. The R-peak detection algorithm was compared with the gold standard Pan-Tompkins, obtaining a p-value > 0.05 on the Fantasia Database, which means that there is no statistical difference between them. Subsequently derived features such as duration, amplitude, subtended area, and P wave slope have been used to discriminate healthy subjects from AF patients. The P-wave slope emerged as the most effective feature, achieving a classification accuracy of 100% and 96% for the training and test sets, respectively. This algorithm thus represents a significant advancement as it achieves a performance comparable to other deterministic methods based on P wave analysis using only one-minute recordings, thereby enabling accurate diagnosis in a shorter time frame.

Signals doi: 10.3390/signals7010002

Authors: Vanel Lazcano Mabel Vega-Rojas Felipe Calderero

Scene depth information is a key component of any robotic mobile application. Range sensors, such as LiDAR, sonar, or radar, capture depth data of a scene. However, the data captured by these sensors frequently presents missing regions or information with a low confidence level. These missing regions in the depth data could be large areas without information, making it difficult to make decisions, for instance, for an autonomous vehicle. Recovering depth data has become a primary activity for computer vision applications. This work proposes and evaluates an interpolation model to infer dense depth maps from a Lab color space reference picture and an incomplete-depth image embedded in a completion pipeline. The complete proposal pipeline comprises convolutional layers and a convex combination of the infinity Laplacian and non-local means model. The proposed model infers dense depth maps by considering depth data and utilizing clues from a color picture of the scene, along with a metric for computing differences between two pixels. The work contributes (i) the convex combination of the two models to interpolate the data, and (ii) the proposal of a class of function suitable for balancing between different models. The obtained results show that the model outperforms similar models in the KITTI dataset and outperforms our previous implementation in the NYU_v2 dataset, dropping the MSE by 34.86%, 3.35%, and 34.42% for 4×, 8×, 16× upsampling tasks, respectively.

Signals doi: 10.3390/signals7010001

Authors: Antonio Santos da Silva Kevin Herman Muraro Gularte Giovanni Almeida Santos Davi Salomão Soares Corrêa Luís Felipe Oliveira de Melo João Paulo Javidi da Costa José Alfredo Ruiz Vargas Daniel Alves da Silva Tai Fei

Autonomous vehicles (AVs) are transforming transportation by improving safety, efficiency, and intelligence through integrated sensing, computing, and communication technologies. However, their growing reliance on Vehicle-to-Everything (V2X) communication exposes them to cybersecurity vulnerabilities, particularly at the physical layer. Among these, jamming attacks represent a critical threat by disrupting wireless channels and compromising message delivery, severely impacting vehicle coordination and safety. This work investigates the robustness of New Radio (NR)-V2X-enabled vehicular systems under jamming conditions through a dual-methodology approach. First, two Cooperative Intelligent Transport System (C-ITS) scenarios standardized by 3GPP—Do Not Pass Warning (DNPW) and Intersection Movement Assist (IMA)—are implemented in the OMNeT++ simulation environment using Simu5G, Veins, and SUMO. The simulations incorporate four types of jamming strategies and evaluate their impact on key metrics such as packet loss, signal quality, inter-vehicle spacing, and collision risk. Second, a complementary laboratory experiment is conducted using AnaPico vector signal generators (a Keysight Technologies brand) and an Anritsu multi-channel spectrum receiver, replicating controlled wireless conditions to validate the degradation effects observed in the simulation. The findings reveal that jamming severely undermines communication reliability in NR-V2X systems, both in simulation and in practice. These findings highlight the urgent need for resilient NR-V2X protocols and countermeasures to ensure the integrity of cooperative autonomous systems in adversarial environments.