Next Issue
Volume 25, December-2
Previous Issue
Volume 25, November-2
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
8.2
3.5
Journals
Sensors
Volume 25
Issue 23
sensors-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Sensors, Volume 25, Issue 23 (December-1 2025) – 302 articles

Cover Story (view full-size image): Wireless sensors are vital for condition monitoring but rely on unsustainable batteries. While vibration energy harvesting offers a promising alternative, current harvesters often neglect high-frequency energy, which is weak but contains abundant information and is further attenuated by destructive interference from multiple sources. To address these limitations, this paper proposes a symmetrical gradient metamaterial beam (SGMB) integrated with multiple piezoelectric patches to enhance multi-band high-frequency energy harvesting and suppress dual-source destructive interference. Simulation and experimental results demonstrate that the SGMB provides multiple enhanced bands within 1000 Hz–3500 Hz and improves the energy harvesting efficiency by over 100 times, representing a breakthrough for self-powered and self-sensing wireless sensors. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
17 pages, 694 KB  
Article
Movable Antenna-Enabled RIS-Assisted Simultaneous Wireless Information and Power Transfer Systems
by Dun Feng, Xuan Zhang, Xiaofan Yu, Xin Wang and Xiaoye Shi
Sensors 2025, 25(23), 7402; https://doi.org/10.3390/s25237402 - 4 Dec 2025
Viewed by 508
Abstract
The integration of movable antenna (MA) and reconfigurable intelligent surfaces (RIS) offers promising potential for enhancing simultaneous wireless information and power transfer (SWIPT) systems. In this paper, we investigate a novel MA-enabled RIS-assisted SWIPT framework, where both RIS and MA are jointly exploited [...] Read more.
The integration of movable antenna (MA) and reconfigurable intelligent surfaces (RIS) offers promising potential for enhancing simultaneous wireless information and power transfer (SWIPT) systems. In this paper, we investigate a novel MA-enabled RIS-assisted SWIPT framework, where both RIS and MA are jointly exploited to provide additional spatial degrees of freedom and reconfigurable propagation channels. Then, we formulate an energy harvesting maximization problem under communication reliability constraints by jointly optimizing the base station beamforming, RIS phase shifts, and MA positions. To tackle the proposed non-convexity problem, an efficient alternating optimization (AO) algorithm is developed, which is based on successive convex approximation (SCA) and second-order Taylor expansion. The obtained simulation outcomes reveal that incorporating MA into RIS-assisted SWIPT systems leads to notable performance gains over both conventional RIS schemes and fixed-antenna benchmarks. Full article
(This article belongs to the Section Internet of Things)
►▼ Show Figures

Figure 1

23 pages, 4209 KB  
Article
Localization of Radio Signal Sources for Situational Awareness Enhancement
by Krzysztof Malon, Paweł Skokowski and Gregor Pavlin
Sensors 2025, 25(23), 7401; https://doi.org/10.3390/s25237401 - 4 Dec 2025
Viewed by 526
Abstract
This article proposes a novel passive localization framework that leverages detection results from existing distributed radio detectors. The intuition behind this solution is to combine positive (signal detected) and negative (signal not detected) detection results with environmental data to refine localization estimates. Its [...] Read more.
This article proposes a novel passive localization framework that leverages detection results from existing distributed radio detectors. The intuition behind this solution is to combine positive (signal detected) and negative (signal not detected) detection results with environmental data to refine localization estimates. Its novelty lies in providing a comprehensive, multi-dimensional framework for cooperative localization that enhances situational awareness by leveraging existing spectrum-monitoring capabilities. The proposed approach provides an additional functionality for a network of nodes monitoring spectral resources. It allows the transmitter’s location to be estimated based on the detection results of individual nodes. The unquestionable advantage of the proposed solution is that it does not require extra equipment or increased monitoring time. The developed method supports broad operational activities, e.g., tracking of authorized and unauthorized entities, and jammer localization. Using the proposed approach, one can increase efficiency in a given operational environment, and jammer localization. Using the proposed approach, one can increase efficiency in a given operational environment and situational awareness in a cognitive radio network. Furthermore, the experimental results of the estimation algorithm for an exemplary urban area indicate the legitimacy of a cooperative approach to the problem. Full article
(This article belongs to the Special Issue Emerging Trends and Challenges in Reconnaissance, Jamming, and Unmanned Vehicles Capabilities)
►▼ Show Figures

Figure 1

21 pages, 2533 KB  
Article
Coverage-Conflict-Aware RFID Reader Placement with Range Adjustment for Complete Tag Coverage in IIoT
by Chien-Fu Cheng and Bo-Yan Liao
Sensors 2025, 25(23), 7400; https://doi.org/10.3390/s25237400 - 4 Dec 2025
Cited by 1 | Viewed by 380
Abstract
Radio Frequency Identification (RFID) is a core enabler of the Industrial Internet of Things (IIoT), yet dense deployments suffer from tag collisions and reader interference that degrade reliability and inflate infrastructure cost. This study proposes a deterministic Reader Deployment Algorithm with Adjustable Reader [...] Read more.
Radio Frequency Identification (RFID) is a core enabler of the Industrial Internet of Things (IIoT), yet dense deployments suffer from tag collisions and reader interference that degrade reliability and inflate infrastructure cost. This study proposes a deterministic Reader Deployment Algorithm with Adjustable Reader range (RDA2R) to achieve full tag coverage with minimal interference and reader usage. The method divides the monitored field into grid units, evaluates tag coverage weights, activates high-weight readers with interference checks, and adaptively adjusts interrogation ranges. Simulation results under random and congregation tag distributions show that RDA2R requires about 46–47% fewer readers than ARLDL and 32–33% fewer than MR2D, while improving average tag coverage per reader by over 30%. These results demonstrate that RDA2R provides a scalable, interference-aware, and cost-efficient deployment strategy for RFID-enabled IIoT environments. Full article
(This article belongs to the Special Issue RFID and Zero-Power Backscatter Sensors)
►▼ Show Figures

Figure 1

29 pages, 4559 KB  
Article
A Novel Data-Driven Multi-Agent Reinforcement Learning Approach for Voltage Control Under Weak Grid Support
by Jiaxin Wu, Ziqi Wang, Ji Han, Qionglin Li, Ran Sun, Chenhao Li, Yuehan Cheng, Bokai Zhou, Jiaming Guo and Bocheng Long
Sensors 2025, 25(23), 7399; https://doi.org/10.3390/s25237399 - 4 Dec 2025
Viewed by 672
Abstract
To address active voltage control in photovoltaic (PV)-integrated distribution networks characterized by weak voltage support conditions, this paper proposes a multi-agent deep reinforcement learning (MADRL)-based coordinated control method for PV clusters. First, the voltage control problem is formulated as a decentralized partially observable [...] Read more.
To address active voltage control in photovoltaic (PV)-integrated distribution networks characterized by weak voltage support conditions, this paper proposes a multi-agent deep reinforcement learning (MADRL)-based coordinated control method for PV clusters. First, the voltage control problem is formulated as a decentralized partially observable Markov decision process (Dec-POMDP), and a centralized training with decentralized execution (CTDE) framework is adopted, enabling each inverter to make independent decisions based solely on local measurements during the execution phase. To balance voltage compliance with energy efficiency, two barrier functions are designed to reshape the reward function, introducing an adaptive penalization mechanism: a steeper gradient in violation region to accelerate voltage recovery to the nominal range, and a gentler gradient in the safe region to minimize excessive reactive regulation and power losses. Furthermore, six representative MADRL algorithms—COMA, IDDPG, MADDPG, MAPPO, SQDDPG, and MATD3—are employed to solve the active voltage control problem of the distribution network. Case studies based on a modified IEEE 33-bus system demonstrate that the proposed framework ensures voltage compliance while effectively reducing network losses. The MADDPG algorithm achieves a Controllability Ratio (CR) of 91.9% while maintaining power loss at approximately 0.0695 p.u., demonstrating superior convergence and robustness. Comparisons with optimal power flow (OPF) and droop control methods confirm that the proposed approach significantly improves voltage stability and energy efficiency under model-free and communication-constrained weak grid conditions. Full article
(This article belongs to the Topic Advanced Strategies for Smart Grid Reliability and Energy Optimization)
►▼ Show Figures

Figure 1

21 pages, 954 KB  
Article
Enhancing DNN Adversarial Robustness via Dual Stochasticity and Geometric Normalization
by Xiang Wu and Gangtao Han
Sensors 2025, 25(23), 7398; https://doi.org/10.3390/s25237398 - 4 Dec 2025
Viewed by 386
Abstract
Deep neural networks (DNNs) have achieved remarkable progress across various domains, yet they remain highly vulnerable to adversarial attacks, which significantly hinder their deployment in safety-critical applications. While stochastic defenses have shown promise, most existing approaches rely on fixed noise injection and fail [...] Read more.
Deep neural networks (DNNs) have achieved remarkable progress across various domains, yet they remain highly vulnerable to adversarial attacks, which significantly hinder their deployment in safety-critical applications. While stochastic defenses have shown promise, most existing approaches rely on fixed noise injection and fail to account for the geometric stability of the decision space. To address these limitations, we introduce a novel framework, which named as Dual Stochasticity and Geometric Normalization (DSGN). Specifically, DSGN incorporates learnable, input-dependent Gaussian noise into both the feature representation and classifier weights, creating a dual-path stochastic modeling mechanism that captures multi-level predictive uncertainty. To enhance decision consistency, both noisy components are projected onto a unit hypersphere via 𝓁2 normalization, constraining the logit space and promoting angular margin separation. This design stabilizes both the representation and decision geometry, leading to more stable decision boundaries and improved robustness. We evaluate the effectiveness of DSGN on several benchmark datasets and CNNs. Our results indicate that DSGN achieves a robust accuracy improvement of approximately 1% to 6% over the state-of-the-arts baseline model under PGD and 1% to 17% improvement under AutoAttack, demonstrating its effectiveness in enhancing adversarial robustness while maintaining high clean accuracy. Full article
(This article belongs to the Special Issue Advances in Security for Emerging Intelligent Systems)
►▼ Show Figures

Figure 1

18 pages, 4876 KB  
Article
Research on the Dynamic Mechanism and Multi-Parameter Collaborative Optimization of a Cantilevered Conveyor Trough in Combine Harvesters for Vibration Suppression
by Qi He, Zhan Su, Pengfei Qian, Zhong Tang, Zhaoming Zhang, Jiahao Shen and Ting Lu
Sensors 2025, 25(23), 7397; https://doi.org/10.3390/s25237397 - 4 Dec 2025
Cited by 1 | Viewed by 347
Abstract
Excessive swing of the cantilevered conveyor trough is a key issue restricting the working efficiency and operational stability of combine harvesters. To suppress its swing, this study established a dynamic model of the conveyor trough to reveal the influence mechanisms of the initial [...] Read more.
Excessive swing of the cantilevered conveyor trough is a key issue restricting the working efficiency and operational stability of combine harvesters. To suppress its swing, this study established a dynamic model of the conveyor trough to reveal the influence mechanisms of the initial angle, overall length, and cylinder pivot length on its swing characteristics. Orthogonal experimental design and multi-factor analysis of variance were employed to systematically analyze the significance of these three factors on swing amplitude, identifying cylinder pivot length as the most dominant factor. Optimization results determined the optimal parameter combination as an initial angle of 48.33°, an overall length of 1.45 m, and a cylinder pivot length of 1.1 m. Field tests verified that this optimized scheme reduces the swing amplitude by 11.62%, with a minimal error of 0.57% between theoretical and measured values, providing a reliable theoretical and experimental basis for the low-vibration design of combine harvester conveying mechanisms. Full article
(This article belongs to the Special Issue Intelligent Sensing Technology for Structural Vibration Control and Non-Destructive Testing)
►▼ Show Figures

Figure 1

24 pages, 4103 KB  
Article
Conformal Swallowing Accelerometry: Reimagining the Acquisition and Characterization of Swallowing Mechano-Acoustic Signals
by Wilson Yiu Shun Lam, Elaine Kwong, Randolph Chi Kin Leung, Chak Hang Lee, Sanjaya Rai and Leo Kwan Lui
Sensors 2025, 25(23), 7396; https://doi.org/10.3390/s25237396 - 4 Dec 2025
Viewed by 418
Abstract
(1) Background: Non-invasive instrumental measurement of swallowing acoustic signals has rested upon the assumptions of signal symmetry and reproducibility along the cervical region and has hence taken the form of single-point acquisition on optimal sites. This study aimed to (i) revisit such assumptions [...] Read more.
(1) Background: Non-invasive instrumental measurement of swallowing acoustic signals has rested upon the assumptions of signal symmetry and reproducibility along the cervical region and has hence taken the form of single-point acquisition on optimal sites. This study aimed to (i) revisit such assumptions by adopting a conformal array of accelerometers, and hence (ii) lay the foundation for the future design of swallowing accelerometry. (2) Methods: Thirteen young healthy individuals, including eight females (mean age ± SD = 24.38 ± 0.92) and five males (mean age ± SD = 24 ± 3.74), were recruited. Swallowing mechano-acoustic signals of repeated swallowing trials were captured using conformal swallowing accelerometry. The peak intensities and frequencies as well as their respective peak times were extracted from six symmetrical and vertically aligned sites. (3) Results: Three-way ANOVAs with repeated measures suggested differences across trials and channels for both peak intensity and frequency. The additional interaction of bolus volume and repeated trials with a small effect size was also indicated in peak frequency. Intra-personal variability was indicated by coefficients of variance of the peak intensity and frequency of higher than 20%, with values varying within the 95% limits of agreement of at least 10 m/s2 and 100 Hz, respectively. However, intra-trial comparisons of contra-lateral peak intensity and frequency also revealed a high degree of variability, with the 95% limits of agreement up to 12 m/s2 and 240 Hz, respectively. On the other hand, the time points of intra-trial peak intensity and frequency showed a high degree agreement, suggesting the possibility of signal asymmetry. (4) Conclusions: The current findings not only confirmed the previous proposal of intra-personal variability but also demonstrated preliminary counterevidence to the longstanding assumption of signal symmetry. Alternatively, the use of conformal swallowing accelerometry is a promising option for the future design and implementation of non-invasive swallowing mechano-acoustic measurements. Full article
(This article belongs to the Special Issue Biomedical Imaging, Sensing and Signal Processing)
►▼ Show Figures

Figure 1

19 pages, 2090 KB  
Article
Towards In-Vehicle Non-Contact Estimation of EDA-Based Arousal with LiDAR
by Jonas Brandstetter, Eva-Maria Knoch and Frank Gauterin
Sensors 2025, 25(23), 7395; https://doi.org/10.3390/s25237395 - 4 Dec 2025
Viewed by 482
Abstract
Driver monitoring systems are increasingly relying on physiological signals to assess cognitive and emotional states for improved safety and user experience. Electrodermal activity (EDA) is a particularly informative biomarker of arousal but is conventionally measured with skin-contact electrodes, limiting its applicability in vehicles. [...] Read more.
Driver monitoring systems are increasingly relying on physiological signals to assess cognitive and emotional states for improved safety and user experience. Electrodermal activity (EDA) is a particularly informative biomarker of arousal but is conventionally measured with skin-contact electrodes, limiting its applicability in vehicles. This work explores the feasibility of non-contact EDA estimation using Light Detection and Ranging (LiDAR) as a novel sensing modality. In a controlled laboratory setup, LiDAR reflection intensity from the forehead was recorded simultaneously with conventional finger-based EDA. Both classification and regression tasks were performed as follows: feature-based machine learning models (e.g., Random Forest and Extra Trees) and sequence-based deep learning models (e.g., CNN, LSTM, and TCN) were evaluated. Results demonstrate that LiDAR signals capture arousal-related changes, with the best regression model (Temporal Convolutional Network) achieving a mean absolute error of 14.6 on the normalized arousal factor scale (–50 to +50) and a correlation of r = 0.85 with ground-truth EDA. While random split validations yielded high accuracy, performance under leave-one-subject-out evaluation highlighted challenges in cross-subject generalization. The algorithms themselves were not the primary research focus but served to establish feasibility of the approach. These findings provide the first proof-of-concept that LiDAR can remotely estimate EDA-based arousal without direct skin contact, addressing a central limitation of current driver monitoring systems. Future research should focus on larger datasets, multimodal integration, and real-world driving validation to advance LiDAR towards practical in-vehicle deployment. Full article
(This article belongs to the Section Vehicular Sensing)
►▼ Show Figures

Figure 1

24 pages, 8415 KB  
Article
Enhancements and On-Site Experimental Study on Fall Detection Algorithm for Students in Campus Staircase
by Ying Lu, Yuze Cui and Liang Yan
Sensors 2025, 25(23), 7394; https://doi.org/10.3390/s25237394 - 4 Dec 2025
Viewed by 382
Abstract
Campus stairwells, characterized by their crowded nature during certain short periods of time, present a high risk for falls that can lead to dangerous stampedes. Accurate fall detection is crucial for preventing such accidents. However, existing research lacks a detection model that balances [...] Read more.
Campus stairwells, characterized by their crowded nature during certain short periods of time, present a high risk for falls that can lead to dangerous stampedes. Accurate fall detection is crucial for preventing such accidents. However, existing research lacks a detection model that balances high precision with lightweight design and lacks on-site experimental validation to assess practical feasibility. This study addresses these gaps by proposing an enhanced fall recognition model based on YOLOv7, validated through on-site experiments. A dataset on campus stairwell falls was established, capturing diverse stairwell personnel behaviors. Four YOLOv7 improvement schemes were proposed, and numerical comparison experiments identified the best-performing model, combining DO-DConv and Slim-Neck modules. This model achieved an average precision (mAP) of 88.1%, 2.41% higher than the traditional YOLOv7, while reducing GFLOPs from 105.2 to 38.2 and cutting training time by 4 h. A field experiment conducted with 22 groups of participants under small-scale populations and varying lighting conditions preliminarily confirmed that the model’s accuracy is within an acceptable range. The experimental results also analyzed the changes in detection confidence across different population sizes and lighting conditions, offering valuable insights for further model improvement and its practical applications. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

18 pages, 3903 KB  
Article
Tolerance Analysis of Test Mass Alignment Errors for Space-Based Gravitational Wave Detection
by Jun Ke, Ruihong Gao, Jinghan Liu, Mengyang Zhao, Ziren Luo, Jia Shen and Peng Dong
Sensors 2025, 25(23), 7393; https://doi.org/10.3390/s25237393 - 4 Dec 2025
Viewed by 329
Abstract
Space-based gravitational wave detection imposes extremely high requirements on displacement measurement accuracy, with its core measurement components being laser interferometers and inertial sensors. The laser interferometers detect gravitational wave signals by measuring the distance between two test masses (TMs) housed within the inertial [...] Read more.
Space-based gravitational wave detection imposes extremely high requirements on displacement measurement accuracy, with its core measurement components being laser interferometers and inertial sensors. The laser interferometers detect gravitational wave signals by measuring the distance between two test masses (TMs) housed within the inertial sensors. Spatial alignment errors of the TMs relative to the laser interferometers can severely degrade the interferometric performance, primarily by significantly amplifying tilt-to-length (TTL) coupling noise and reducing interferometric efficiency. This paper presents a systematic analysis of the coupling mechanisms between TM alignment errors and TTL coupling noise. We first establish a comprehensive TTL noise model that accounts for alignment errors, then verify and analyze it through optical simulations. This research ultimately clarifies the coupling mechanisms of TM alignment errors in the context of space-borne gravitational wave missions and determines the allowable alignment tolerance specifications required to meet the gravitational wave detection sensitivity requirements. This work provides critical theoretical foundations and design guidance for the ground alignment procedures and on-orbit performance prediction of future space-based gravitational wave detection missions. Full article
(This article belongs to the Section Optical Sensors)
►▼ Show Figures

Figure 1

26 pages, 30428 KB  
Article
Lightweight and Compact Pulse Radar for UAV Platforms for Mid-Air Collision Avoidance
by Dawid Sysak, Arkadiusz Byndas, Tomasz Karas and Grzegorz Jaromi
Sensors 2025, 25(23), 7392; https://doi.org/10.3390/s25237392 - 4 Dec 2025
Viewed by 546
Abstract
Small and medium Unmanned Aerial Vehicles (UAVs) are commonly equipped with diverse sensors for situational awareness, including cameras, Frequency-Modulated Continuous-Wave (FMCW) radars, Light Detection and Ranging (LiDAR) systems, and ultrasonic sensors. However, optical systems are constrained by adverse weather and darkness, while the [...] Read more.
Small and medium Unmanned Aerial Vehicles (UAVs) are commonly equipped with diverse sensors for situational awareness, including cameras, Frequency-Modulated Continuous-Wave (FMCW) radars, Light Detection and Ranging (LiDAR) systems, and ultrasonic sensors. However, optical systems are constrained by adverse weather and darkness, while the limited detection range of compact FMCW radars-typically a few hundred meters-is often insufficient for higher-speed UAVs, particularly those operating Beyond Visual Line of Sight (BVLOS). This paper presents a Collision Avoidance System (CAS) based on a lightweight pulse radar, targeting medium UAV platforms (10–300 kg MTOM) where installing large, nose-mounted radars is impractical. The system is designed for obstacle detection at ranges of 1–3 km, directly addressing the standoff distance limitations of conventional sensors. Beyond its primary sensing function, the pulse architecture offers several operational advantages. Its lower time-averaged power also results in a reduced electromagnetic footprint, mitigating interference and supporting emission-control objectives. Furthermore, pulse radar offers greater robustness against interference in dense electromagnetic environments and lower power consumption, both of which directly enhance UAV operational endurance. Field tests demonstrated a one-to-one correspondence between visually identified objects and radar detections across 1–3 km, with PFA = 1.5%, confirming adequate standoff for tens of seconds of maneuvering time, with range resolution of 3.75 m and average system power below 80 W. Full article
(This article belongs to the Section Radar Sensors)
►▼ Show Figures

Figure 1

25 pages, 4787 KB  
Article
Implementation of Vital Signs Detection Algorithm for Supervising the Evacuation of Individuals with Special Needs
by Krzysztof Konopko, Dariusz Janczak and Wojciech Walendziuk
Sensors 2025, 25(23), 7391; https://doi.org/10.3390/s25237391 - 4 Dec 2025
Viewed by 351
Abstract
The article describes a system for monitoring the vital parameters of evacuated individuals, integrating three key functionalities: pulse detection, verification of wristband contact with the skin, and motion recognition. For pulse detection, the system employs the MAX30102 optical sensor and a signal processing [...] Read more.
The article describes a system for monitoring the vital parameters of evacuated individuals, integrating three key functionalities: pulse detection, verification of wristband contact with the skin, and motion recognition. For pulse detection, the system employs the MAX30102 optical sensor and a signal processing algorithm presented in the study. The algorithm is based on spectral analysis using the Fast Fourier Transform (FFT) and incorporates a nonparametric estimator of the probability density function (PDF) in the form of Kernel Density Estimation (KDE). This developed real-time algorithm enables reliable assessment of vital parameters of evacuated individuals. The wristband contact with the skin is verified by measuring the brightness of backscattered light and the temperature of the wrist. Motion detection is achieved using the MPU-9250 inertial module, which analyzes acceleration across three axes. This allows the system to distinguish between states of rest and physical activity, which is crucial for accurately interpreting vital parameters during evacuation. The experimental studies, which were performed on a representative group of individuals, confirmed the correctness of the developed algorithm. The system ensures reliable monitoring of vital parameters by combining precise pulse detection, skin contact verification, and motion analysis. The classifier achieves nearly 95% accuracy and an F1-score of 0.9465, which indicates its high quality. This level of effectiveness can be considered fully satisfactory for evacuation monitoring systems. Full article
(This article belongs to the Special Issue Sensing Signals for Biomedical Monitoring—2nd Edition)
►▼ Show Figures

Figure 1

14 pages, 2827 KB  
Article
Accelerometer-Based Gait Analysis as a Predictive Tool for Mild Cognitive Impairment in Older Adults
by Junwei Shen, Yoshiko Nagata, Toshiya Shimamoto, Shigehito Matsubara, Masato Nakamura, Fumiya Sato, Takuya Motoshima, Katsuhisa Uchino, Akira Mori, Miwa Nogami, Yuki Harada, Makoto Uchino and Shinichiro Nakamura
Sensors 2025, 25(23), 7390; https://doi.org/10.3390/s25237390 - 4 Dec 2025
Viewed by 559
Abstract
This study explores the potential of accelerometer-based gait analysis as a non-invasive approach for predicting cognitive impairment in older adults. A total of 75 participants (61.3% female; mean age: 78.9 years), including cognitively normal individuals and patients with dementia, were enrolled. Walking data [...] Read more.
This study explores the potential of accelerometer-based gait analysis as a non-invasive approach for predicting cognitive impairment in older adults. A total of 75 participants (61.3% female; mean age: 78.9 years), including cognitively normal individuals and patients with dementia, were enrolled. Walking data were collected using a six-axis waist-worn accelerometer during self-paced locomotion. Allan variance (AVAR), a robust statistical measure of frequency stability, was applied to characterize gait dynamics. AVAR-derived features, combined with participant age, were used as inputs to machine learning models, logistic regression and Light Gradient Boosting Machine (LightGBM) for classifying cognitive status based on Mini-Mental State Examination (MMSE) scores. LightGBM achieved superior performance (AUC = 0.92) compared to logistic regression (AUC = 0.85). Although mild cognitive impairment (MCI) cases were grouped with cognitively normal participants, gait-based classification revealed that MCI individuals exhibited patterns more similar to those with cognitive impairment. These results suggest that AVAR-based gait features are promising for early detection of cognitive decline in older adults. Full article
(This article belongs to the Section Wearables)
►▼ Show Figures

Graphical abstract

24 pages, 3243 KB  
Article
A State-Space Framework for Parallelizing Digital Signal Processing in Coherent Optical Receivers
by Jinyang Wang, Zhugang Wang and Di Liu
Sensors 2025, 25(23), 7389; https://doi.org/10.3390/s25237389 - 4 Dec 2025
Viewed by 416
Abstract
Ultra-high sampling rates in coherent optical front-ends increasingly exceed the processing capabilities of real-time baseband processors, creating a bottleneck in coherent free-space optical communication systems. We propose a unified state-space framework to systematically parallelize digital signal processing (DSP) algorithms. This approach transforms an [...] Read more.
Ultra-high sampling rates in coherent optical front-ends increasingly exceed the processing capabilities of real-time baseband processors, creating a bottleneck in coherent free-space optical communication systems. We propose a unified state-space framework to systematically parallelize digital signal processing (DSP) algorithms. This approach transforms an algorithm’s transfer function into a state-space representation from which a parallel architecture is derived through matrix operations, overcoming the complexity of traditional ad hoc methods. Crucially, our framework enables an analysis of parallelization-induced latency. We introduce the parallel equivalent delay (PED) metric and demonstrate that it introduces right-half-plane zeros into the loop’s transfer function, thereby fundamentally constraining stability. This analysis leads to the derivation of “Throughput–Bandwidth Product” (TBP), a constant that provides a design guideline linking maximum stable loop bandwidth to the parallelization factor. The framework’s efficacy is demonstrated by designing a parallel Costas carrier recovery loop. Simulations validate its performance, confirm the TBP limit, and show significant advantages over conventional feedforward estimators, especially in low-SNR conditions. Implementation results on a AMD XCVU13P FPGA demonstrate that the proposed 50-parallel architecture achieves a throughput of 15.625 Gsps at a clock frequency of 312.5 MHz with a logic utilization below 7%. The experimental results confirm the theoretical trade-off between throughput and loop bandwidth, verifying the proposed design methodology. Full article
(This article belongs to the Section Communications)
►▼ Show Figures

Figure 1

14 pages, 2590 KB  
Article
Multifractal Cascade Modeling Reveals Fundamental Limits of Current Neuroimaging Strategies
by Madhur Mangalam
Sensors 2025, 25(23), 7388; https://doi.org/10.3390/s25237388 - 4 Dec 2025
Viewed by 330
Abstract
Neuroimaging assumes spatial and temporal uniformity, yet brain activity exhibits a multifractal cascade structure with intermittent bursts and long-range dependencies. We use controlled simulations to test how well standard sampling strategies (random, grid-based, hierarchical; N = 10–2000 sensors) recover [...] Read more.
Neuroimaging assumes spatial and temporal uniformity, yet brain activity exhibits a multifractal cascade structure with intermittent bursts and long-range dependencies. We use controlled simulations to test how well standard sampling strategies (random, grid-based, hierarchical; N = 10–2000 sensors) recover statistical properties—mean, variability, burstiness, and fractal dimension—from synthetic multifractal brain fields. Estimation errors deviate substantially from the classical N−1/2 scaling expected under independent sampling. For higher-order statistics like burstiness, error reduction is remarkably flat in log–log space: orders-of-magnitude increases in sensor density yield virtually no improvement. Grid sampling performs best for fractal dimension at high densities; hierarchical sampling is more stable for burstiness. These results indicate that current neuroimaging fundamentally underestimates brain complexity and variability, with major implications for interpreting both healthy and pathological brain function. Full article
(This article belongs to the Special Issue Advanced Sensors in Brain–Computer Interfaces)
►▼ Show Figures

Figure 1

28 pages, 3146 KB  
Article
Predicting the Lifespan of Twisted String Actuators Using Empirical and Hybrid Machine Learning Approaches
by Hai Nguyen, Chanthol Eang and Seungjae Lee
Sensors 2025, 25(23), 7387; https://doi.org/10.3390/s25237387 - 4 Dec 2025
Viewed by 430
Abstract
Predicting the fatigue lifespan of Twisted String Actuators (TSAs) is essential for improving the reliability of robotic and mechanical systems that rely on flexible transmission mechanisms. Traditional empirical approaches based on regression or Weibull distribution analysis have provided useful approximations, yet they often [...] Read more.
Predicting the fatigue lifespan of Twisted String Actuators (TSAs) is essential for improving the reliability of robotic and mechanical systems that rely on flexible transmission mechanisms. Traditional empirical approaches based on regression or Weibull distribution analysis have provided useful approximations, yet they often struggle to capture nonlinear dependencies and stochastic influences inherent to real-world fatigue behavior. This study introduces and compares four machine learning (ML) models—Linear Regression, Random Forest, XGBoost, and Gaussian Process Regression (GPR)—for predicting TSA lifespan under varying weight (W), number of strings (N), and diameter (D) conditions. Building upon this comparison, a hybrid physics-guided model is proposed by integrating an empirical fatigue life equation with an XGBoost residual-correction model. Experimental data collected from repetitive actuation tests (144 valid samples) served as the basis for training and validation. The hybrid model achieved an R2 = 0.9856, RMSE = 5299.47 cycles, and MAE = 3329.67 cycles, outperforming standalone ML models in cross-validation consistency (CV R2 = 0.9752). The results demonstrate that physics-informed learning yields superior interpretability and generalization even in limited-data regimes. These findings highlight the potential of hybrid empirical–ML modeling for component life prediction in robotic actuation systems, where experimental fatigue data are scarce and operating conditions vary. Full article
(This article belongs to the Collection Robotics, Sensors and Industry 4.0)
►▼ Show Figures

Figure 1

14 pages, 2978 KB  
Article
Simulation and Experiment of Tilted Fiber Bragg Grating Humidity Sensor Coated with PVA/GO Nanofiber Films by Electrospinning
by Li Deng, Hao Sun, Jiawei Xi, Yanxin Yang, Xin Liu, Chaochao Jian, Xiang Li and Jinze Li
Sensors 2025, 25(23), 7386; https://doi.org/10.3390/s25237386 - 4 Dec 2025
Viewed by 387
Abstract
Relative humidity (RH) and temperature are crucial parameters in environmental monitoring and have attracted significant attention. However, traditional commercial sensors typically suffer from inherent limitations such as structural complexity, bulkiness, and high manufacturing costs. To address these issues, this study proposes a novel [...] Read more.
Relative humidity (RH) and temperature are crucial parameters in environmental monitoring and have attracted significant attention. However, traditional commercial sensors typically suffer from inherent limitations such as structural complexity, bulkiness, and high manufacturing costs. To address these issues, this study proposes a novel tilted fiber Bragg grating (TFBG)-based optical fiber humidity sensor, coated with a composite film of polyvinyl alcohol (PVA) and graphene oxide (GO). First, the sensing mechanisms of the TFBG functionalized with nanofiber films were theoretically analyzed, and the transmission spectra of TFBG under varied structural parameters were simulated. These theoretical investigations laid a solid foundation for subsequent experimental validation. Subsequently, PVA/GO composite nanofiber films tailored for humidity sensing were fabricated by electrospinning technology, and the proposed TFBG sensor was experimentally implemented in accordance with the theoretical design. The experimental results indicate that the developed sensor exhibits a humidity sensitivity of −0.24 pm/%RH within the RH range of 35–85%. Furthermore, we calculated temperature and RH changes while discounting cross-sensitivity, thereby enabling simultaneous decoupling of temperature and RH measurements. Owing to its distinctive advantages of compact size, light weight, and cost-effectiveness, the proposed TFBG sensor holds great promise for practical applications in environmental monitoring. Full article
(This article belongs to the Section Optical Sensors)
►▼ Show Figures

Figure 1

20 pages, 4697 KB  
Article
A Review of Video-Based Monitoring Systems for Geohazard Early Warning
by Haoran Dong, Shuzhong Sheng and Chong Xu
Sensors 2025, 25(23), 7385; https://doi.org/10.3390/s25237385 - 4 Dec 2025
Viewed by 583
Abstract
In recent years, video-based monitoring systems have been widely adopted across multiple domains and have become particularly vital in geohazard monitoring and early warning. These systems overcome the inherent limitations of conventional monitoring techniques by enabling real-time, non-contact, and intuitive visual observation of [...] Read more.
In recent years, video-based monitoring systems have been widely adopted across multiple domains and have become particularly vital in geohazard monitoring and early warning. These systems overcome the inherent limitations of conventional monitoring techniques by enabling real-time, non-contact, and intuitive visual observation of geologically hazardous sites. With the integration of machine learning and other advanced analytical methods, video-based systems can process and interpret image data in real time, thereby supporting rapid detection and timely early warning of potential geohazards. This substantially improves both the efficiency and accuracy of monitoring efforts. Drawing on domestic and international research, this article provides a comprehensive review of video-based monitoring technologies, machine learning–driven video image processing, and multi-source data fusion approaches. It systematically summarizes their underlying technical principles and applications in geohazard monitoring and early warning, and offers an in-depth analysis of their practical advantages and future development trends. This review aims to serve as a valuable reference for advancing research and innovation in this field. Full article
(This article belongs to the Section Environmental Sensing)
►▼ Show Figures

Figure 1

26 pages, 729 KB  
Article
Sensor-Based Cyber Risk Management in Railway Infrastructure Under the NIS2 Directive
by Rafał Wachnik, Katarzyna Chruzik and Bolesław Pochopień
Sensors 2025, 25(23), 7384; https://doi.org/10.3390/s25237384 - 4 Dec 2025
Viewed by 471
Abstract
This study introduces a sensor-centric cybersecurity framework for railway infrastructure that extends Failure Mode and Effects Analysis (FMEA) from traditional reliability evaluation into the domain of cyber-induced failures affecting data integrity, availability and authenticity. The contribution lies in bridging regulatory obligations of the [...] Read more.
This study introduces a sensor-centric cybersecurity framework for railway infrastructure that extends Failure Mode and Effects Analysis (FMEA) from traditional reliability evaluation into the domain of cyber-induced failures affecting data integrity, availability and authenticity. The contribution lies in bridging regulatory obligations of the NIS2 Directive with field-layer monitoring by enabling risk indicators to evolve dynamically rather than remain static documentation artefacts. The approach is demonstrated using a scenario-based dataset collected from approximately 250 trackside, rolling-stock, environmental and power-monitoring sensors deployed over a 25 km operational segment, with representative anomalies generated through controlled spoofing, replay and injection conditions. Risk was evaluated using RPN scores derived from Severity–Occurrence–Detectability scales, while anomaly-detection performance was observed through detection-latency variation, changes in RPN distribution, and qualitative responsiveness of timestamp-based alerts. Instead of presenting a fixed benchmark, the results show how evidence from real sensor streams can recalibrate O and D factors in near-real-time and reduce undetected exposure windows, enabling measurable compliance documentation aligned with NIS2 Article 21. The findings confirm that coupling FMEA with streaming telemetry creates a verifiable risk-evaluation loop and supports a transition toward continuous, evidence-driven cybersecurity governance in railway systems. Full article
(This article belongs to the Section Internet of Things)
►▼ Show Figures

Figure 1

21 pages, 5877 KB  
Article
High-Resolution Low-Sidelobe Waveform Design Based on HFPFM Coding Model for SAR
by Yu Gao, Guodong Jin, Xifeng Zhang and Daiyin Zhu
Sensors 2025, 25(23), 7383; https://doi.org/10.3390/s25237383 - 4 Dec 2025
Viewed by 332
Abstract
Radar waveform design is an important approach to radar system performance enhancement. For a long time, synthetic aperture radar (SAR) systems have utilized linear frequency modulation (LFM) waveforms as transmitted signals and have relied on window functions to suppress sidelobes. However, this approach [...] Read more.
Radar waveform design is an important approach to radar system performance enhancement. For a long time, synthetic aperture radar (SAR) systems have utilized linear frequency modulation (LFM) waveforms as transmitted signals and have relied on window functions to suppress sidelobes. However, this approach significantly degrades system signal-to-noise ratio (SNR) and resolution. Nonlinear frequency modulation (NLFM) waveforms can suppress sidelobes without SNR loss and have been widely applied in the SAR field in recent years. Nonetheless, they still cannot completely avoid resolution loss. To address this, this article, based on an advanced High-Freedom Parameterized Frequency Modulation (HFPFM) coding model, constructs a waveform sidelobe optimization model constrained by mainlobe widening and solves it using a gradient descent method. Through detailed experiments, we found that the optimized waveform, compared to the LFM waveform, can reduce sidelobes by more than 9 dB without widening the mainlobe, thereby simultaneously avoiding the resolution and SNR losses caused by window function weighting. In addition, this optimization method can efficiently and rapidly optimize all parameters simultaneously using only matrix multiplication and fast Fourier transform (FFT)/inverse fast Fourier transform (IFFT). The SAR point target imaging simulation results verify that the optimized waveform can clearly image weak targets near strong targets, which proves the effectiveness of the proposed method. Full article
(This article belongs to the Special Issue SAR Imaging Technologies and Applications)
►▼ Show Figures

Figure 1

12 pages, 795 KB  
Article
Intraocular Cytokine Level Prediction from Fundus Images and Optical Coherence Tomography
by Hidenori Takahashi, Taiki Tsuge, Yusuke Kondo, Yasuo Yanagi, Satoru Inoda, Shohei Morikawa, Yuki Senoo, Toshikatsu Kaburaki, Tetsuro Oshika and Toshihiko Yamasaki
Sensors 2025, 25(23), 7382; https://doi.org/10.3390/s25237382 - 4 Dec 2025
Viewed by 414
Abstract
The relationship between retinal images and intraocular cytokine profiles remains largely unexplored, and no prior work has systematically compared fundus- and OCT-based deep learning models for cytokine prediction. We aimed to predict intraocular cytokine concentrations using color fundus photographs (CFP) and retinal optical [...] Read more.
The relationship between retinal images and intraocular cytokine profiles remains largely unexplored, and no prior work has systematically compared fundus- and OCT-based deep learning models for cytokine prediction. We aimed to predict intraocular cytokine concentrations using color fundus photographs (CFP) and retinal optical coherence tomography (OCT) with deep learning. Our pipeline consisted of image preprocessing, convolutional neural network–based feature extraction, and regression modeling for each cytokine. Deep learning was implemented using AutoGluon, which automatically explored multiple architectures and converged on ResNet18, reflecting the small dataset size. Four approaches were tested: (1) CFP alone, (2) CFP plus demographic/clinical features, (3) OCT alone, and (4) OCT plus these features. Prediction performance was defined as the mean coefficient of determination (R2) across 34 cytokines, and differences were evaluated using paired two-tailed t-tests. We used data from 139 patients (152 eyes) and 176 aqueous humor samples. The cohort consisted of 85 males (61%) with a mean age of 73 (SD 9.8). Diseases included 64 exudative age-related macular degeneration, 29 brolucizumab-associated endophthalmitis, 19 cataract surgeries, 15 retinal vein occlusion, and 8 diabetic macular edema. Prediction performance was generally poor, with mean R2 values below zero across all approaches. The CFP-only model (–0.19) outperformed CFP plus demographics (–24.1; p = 0.0373), and the OCT-only model (–0.18) outperformed OCT plus demographics (–14.7; p = 0.0080). No significant difference was observed between CFP and OCT (p = 0.9281). Notably, VEGF showed low predictability (31st with CFP, 12th with OCT). Full article
(This article belongs to the Section Optical Sensors)
►▼ Show Figures

Figure 1

28 pages, 2836 KB  
Article
MA-EVIO: A Motion-Aware Approach to Event-Based Visual–Inertial Odometry
by Mohsen Shahraki, Ahmed Elamin and Ahmed El-Rabbany
Sensors 2025, 25(23), 7381; https://doi.org/10.3390/s25237381 - 4 Dec 2025
Viewed by 575
Abstract
Indoor localization remains a challenging task due to the unavailability of reliable global navigation satellite system (GNSS) signals in most indoor environments. One way to overcome this challenge is through visual–inertial odometry (VIO), which enables real-time pose estimation by fusing camera and inertial [...] Read more.
Indoor localization remains a challenging task due to the unavailability of reliable global navigation satellite system (GNSS) signals in most indoor environments. One way to overcome this challenge is through visual–inertial odometry (VIO), which enables real-time pose estimation by fusing camera and inertial measurements. However, VIO suffers from performance degradation under high-speed motion and in poorly lit environments. In such scenarios, motion blur, sensor noise, and low temporal resolution reduce the accuracy and robustness of the estimated trajectory. To address these limitations, we propose a motion-aware event-based VIO (MA-EVIO) system that adaptively fuses asynchronous event data, frame-based imagery, and inertial measurements for robust and accurate pose estimation. MA-EVIO employs a hybrid tracking strategy combining sparse feature matching and direct photometric alignment. A key innovation is its motion-aware keyframe selection, which dynamically adjusts tracking parameters based on real-time motion classification and feature quality. This motion awareness also enables adaptive sensor fusion: during fast motion, the system prioritizes event data, while under slow or stable motion, it relies more on RGB frames and feature-based tracking. Experimental results on the DAVIS240c and VECtor benchmarks demonstrate that MA-EVIO outperforms state-of-the-art methods, achieving a lower mean position error (MPE) of 0.19 on DAVIS240c compared to 0.21 (EVI-SAM) and 0.24 (PL-EVIO), and superior performance on VECtor with MPE/mean rotation error (MRE) of 1.19%/1.28 deg/m versus 1.27%/1.42 deg/m (EVI-SAM) and 1.93%/1.56 deg/m (PL-EVIO). These results validate the effectiveness of MA-EVIO in challenging dynamic indoor environments. Full article
(This article belongs to the Special Issue Multi-Sensor Integration for Mobile and UAS Mapping)
►▼ Show Figures

Figure 1

24 pages, 3536 KB  
Article
Practical Predefined-Time Sliding-Mode Adaptive Resilient Control for PMSM Cyber–Physical Systems
by Zhenzhong Wang, Shu Zhang, Yun Jiang and Chunwu Yin
Sensors 2025, 25(23), 7380; https://doi.org/10.3390/s25237380 - 4 Dec 2025
Viewed by 342
Abstract
The permanent magnet synchronous motor (PMSM) is extensively utilized in the power drive systems of Cyber–Physical Systems (CPSs). In scenarios where control signals are subjected to malicious attacks within the network, ensuring that the PMSM achieves its designated speed within a specified timeframe [...] Read more.
The permanent magnet synchronous motor (PMSM) is extensively utilized in the power drive systems of Cyber–Physical Systems (CPSs). In scenarios where control signals are subjected to malicious attacks within the network, ensuring that the PMSM achieves its designated speed within a specified timeframe serves as a critical metric for evaluating the efficacy of security control strategies in networked systems. To address practical challenges arising from updates to controlled objects at the physical layer and limitations of control layer algorithms—wherein convergence time for system trajectory tracking errors (TTEors) may extend indefinitely—we have developed a novel resilient control algorithm with predefined-time convergence (PreTC) tailored for uncertain PMSMs susceptible to cyber threats. Firstly, we introduce an innovative Lyapunov stability criterion characterized by an adjustable gain reaching law alongside PreTC. Following this, we design an SMS (SMS) that incorporates PreTC and employ an extreme learning machine (ELM) to facilitate real-time identification of both physical layer models and malicious cyber-attacks. A sliding-mode adaptive resilient controller devoid of explicit physical model information is proposed for CPSs, with Lyapunov stability theory substantiating the system’s predefined-time (PDT) stability. This significantly enhances resilience against malicious cyber-attacks and other uncertainties. Finally, comparative simulations involving four distinct resilient control algorithms demonstrate that our proposed algorithm not only guarantees predetermined convergence times but also exhibits robust resistance to cyber-attacks, parameter perturbations, and external disturbances—notably achieving a motor speed tracking error accuracy of 0.008. These findings validate the superior robustness and effectiveness of our control algorithm against malicious cyber threats. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

22 pages, 1188 KB  
Article
EFDepth: A Monocular Depth Estimation Model for Multi-Scale Feature Optimization
by Fengchun Liu, Xinying Shao, Chunying Zhang, Liya Wang, Lu Liu and Jing Ren
Sensors 2025, 25(23), 7379; https://doi.org/10.3390/s25237379 - 4 Dec 2025
Viewed by 526
Abstract
To address the accuracy issues in monocular depth estimation caused by insufficient feature extraction and inadequate context modeling, a multi-scale feature optimization model named EFDepth was proposed to improve prediction performance. This framework adopted an encoder–decoder structure: the encoder (EC-Net) was composed of [...] Read more.
To address the accuracy issues in monocular depth estimation caused by insufficient feature extraction and inadequate context modeling, a multi-scale feature optimization model named EFDepth was proposed to improve prediction performance. This framework adopted an encoder–decoder structure: the encoder (EC-Net) was composed of MobileNetV3-E and ETFBlock, and its features were optimized through multi-scale dilated convolution; the decoder (LapFA-Net) combined the Laplacian pyramid and the FMA module to enhance cross-scale feature fusion and output accurate depth maps. Comparative experiments between EFDepth and algorithms including Lite-mono, Hr-depth, and Lapdepth were conducted on the KITTI datasets. The results show that, for the three error metrics—RMSE (Root Mean Square Error), AbsRel (Absolute Relative Error), and SqRel (Squared Relative Error)—EFDepth is 1.623, 0.030, and 0.445 lower than the average values of the comparison algorithms, respectively, and for the three accuracy metrics, it is 0.052, 0.023, and 0.011 higher than the average values of the comparison algorithms, respectively. Experimental results indicate that EFDepth outperforms the comparison methods in most metrics, providing an effective reference for monocular depth estimation and 3D reconstruction of complex scenes. Full article
(This article belongs to the Section Sensing and Imaging)
►▼ Show Figures

Figure 1

20 pages, 824 KB  
Article
MAGTF-Net: Dynamic Speech Emotion Recognition with Multi-Scale Graph Attention and LLD Feature Fusion
by Shiyin Zhu, Yinggang Xie and Zhiliang Wang
Sensors 2025, 25(23), 7378; https://doi.org/10.3390/s25237378 - 4 Dec 2025
Viewed by 399
Abstract
In this paper, we propose a novel speech emotion recognition model, named MAGTF-Net (Multi-scale Attention Graph Transformer Fusion Network), which addresses the challenges faced by traditional hand-crafted feature-based approaches in modeling complex emotional nuances and dynamic contextual dependencies. Although existing state-of-the-art methods have [...] Read more.
In this paper, we propose a novel speech emotion recognition model, named MAGTF-Net (Multi-scale Attention Graph Transformer Fusion Network), which addresses the challenges faced by traditional hand-crafted feature-based approaches in modeling complex emotional nuances and dynamic contextual dependencies. Although existing state-of-the-art methods have achieved improvements in recognition performance, they often fail to simultaneously capture both local acoustic features and global temporal structures, and they lack adaptability to variable-length speech utterances, thereby limiting their accuracy and robustness in recognizing complex emotional expressions. To tackle these challenges, we design a log-Mel spectrogram feature extraction branch that combines a Multi-scale Attention Graph (MAG) structure with a Transformer encoder, where the Transformer module adaptively performs dynamic modeling of speech sequences with varying lengths. In addition, a low-level descriptor (LLD) feature branch is introduced, where a multilayer perceptron (MLP) is employed for complementary feature modeling. The two feature branches are fused and subsequently classified through a fully connected layer, further enhancing the expressive capability of emotional representations. Moreover, a label-smoothing-enhanced cross-entropy loss function is adopted to improve the model’s recognition performance on difficult-to-classify emotional categories. Experiments conducted on the IEMOCAP dataset demonstrate that MAGTF-Net achieves weighted accuracy (WA) and unweighted accuracy (UA) scores of 69.15% and 70.86%, respectively, outperforming several baseline models. Further ablation studies validate the significant contributions of each module in the Mel-spectrogram branch and the LLD feature branch to the overall performance improvement. The proposed method effectively integrates local, global, and multi-source feature information, significantly enhancing the recognition of complex emotional expressions and providing new theoretical and practical insights for the field of speech emotion recognition. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

17 pages, 3211 KB  
Article
From Static to Dynamic: Complementary Roles of FSR and Piezoelectric Sensors in Wearable Gait and Pressure Monitoring
by Sara Sêco, Vítor Miguel Santos, Sara Valvez, Beatriz Branquinho Gomes, Maria Augusta Neto and Ana Martins Amaro
Sensors 2025, 25(23), 7377; https://doi.org/10.3390/s25237377 - 4 Dec 2025
Viewed by 560
Abstract
Objective: Plantar pressure abnormalities have a significant impact on mobility and quality of life. Real-time pressure monitoring is essential in clinical and rehabilitation settings for assessing patient progress and refining treatment protocols. Instrumental and particularly smart insoles offer a promising solution by collecting [...] Read more.
Objective: Plantar pressure abnormalities have a significant impact on mobility and quality of life. Real-time pressure monitoring is essential in clinical and rehabilitation settings for assessing patient progress and refining treatment protocols. Instrumental and particularly smart insoles offer a promising solution by collecting biomechanical data during daily activities. However, determining the optimal combination of sensor type, number, and placement remains a key challenge for ensuring accurate and reliable measurements. This study proposes a methodology for identifying the most appropriate sensor technology for wearable insoles, with a focus on data accuracy, system efficiency, and practical applicability. Additionally, it examines the correlation between sensor signals and material behavior during compression testing. Methods: Two insole prototypes underwent compression testing: one equipped with a Force Sensitive Resistor (FSR) sensor and one with a piezoelectric sensor, both positioned at the heel. Three trials per prototype assessed consistency and repeatability. Real-time data acquisition utilized a microcontroller system, and signals were processed using a sixth-order Butterworth low-pass filter with a 5 Hz cutoff frequency to reduce noise. Results: FSR sensors demonstrated stable static responses but saturated rapidly beyond 20 N, with performance degradation observed after repeated loading cycles. Piezoelectric sensors exhibited excellent dynamic sensitivity with sharp voltage peaks but proved unable to measure sustained static pressure. Conclusions: FSR sensors are well-suited for static postural assessment and continuous pressure monitoring, while piezoelectric sensors excel in dynamic gait analysis. This comparative framework establishes a foundation for developing future smart insole systems that deliver accurate, real-time rehabilitation monitoring. Full article
(This article belongs to the Special Issue Sensor Systems for Gesture Recognition (3rd Edition))
►▼ Show Figures

Figure 1

13 pages, 2181 KB  
Article
Association Between Stride Parameters and Racetrack Curvature for Thoroughbred Chuckwagon Horses
by Matthijs van den Broek, Zoe Y. S. Chan, Charlotte De Bruyne, Karelhia Garcia-Alamo, Sara Skotarek Loch and Thilo Pfau
Sensors 2025, 25(23), 7376; https://doi.org/10.3390/s25237376 - 4 Dec 2025
Viewed by 477
Abstract
Increased risk of musculoskeletal injury in galloping racehorses has been linked to decreased stride length and reduced speed over consecutive races prior to the injury. As racetrack curvature influences horses’ maximal speed, we hypothesized it also affects stride parameters. During training sessions, twenty-eight [...] Read more.
Increased risk of musculoskeletal injury in galloping racehorses has been linked to decreased stride length and reduced speed over consecutive races prior to the injury. As racetrack curvature influences horses’ maximal speed, we hypothesized it also affects stride parameters. During training sessions, twenty-eight wagon-pulling Thoroughbred Chuckwagon horses were equipped with Global Navigation Satellite System (GNSS) loggers, allowing for identification of speed, stride length (SL) and stride frequency (SF), and average speed, SL and SF were calculated for consecutive 100 m sections. Effects of curvature on speed were investigated with a linear mixed model with speed as output variable, curvature as fixed factor, and horse as random factor. Effects of curvature and speed on stride parameters were investigated with linear mixed models with output variables SL and SF, continuous covariates speed, curvature, and the two-way interaction between curvature and speed as fixed factors, and horse as random factor. Curvature was associated with a significant increase in speed (p = 0.004), decrease in SL (p < 0.001) and increase in SF (p < 0.001), and for SL and SF the magnitude of these effects was dependent on speed (p < 0.001). At a curvature of 60° per 100 m, an increase in speed of 0.264 m/s was found compared to the straight, although this effect is likely confounded by fatigue. At the median speed of 14.5 m/s and a curvature of 60° per 100 m, a SF increase of 0.053 Hz (+2.4%) and a SL reduction of 0.137 m (−2.1%) was found compared to the straight. This is in the same order of magnitude as the 0.10 m SL reduction over consecutive races previously associated with increased injury risk. We conclude that, in Chuckwagon horses, interactions between speed and curvature are affecting stride parameters that have previously been identified as predictors of musculoskeletal injuries. Full article
(This article belongs to the Section Navigation and Positioning)
►▼ Show Figures

Figure 1

17 pages, 3220 KB  
Article
ArecaNet: Robust Facial Emotion Recognition via Assembled Residual Enhanced Cross-Attention Networks for Emotion-Aware Human–Computer Interaction
by Jaemyung Kim and Gyuho Choi
Sensors 2025, 25(23), 7375; https://doi.org/10.3390/s25237375 - 4 Dec 2025
Viewed by 438
Abstract
Recently, the convergence of advanced sensor technologies and innovations in artificial intelligence and robotics has highlighted facial emotion recognition (FER) as an essential component of human–computer interaction (HCI). Traditional FER studies based on handcrafted features and shallow machine learning have shown a limited [...] Read more.
Recently, the convergence of advanced sensor technologies and innovations in artificial intelligence and robotics has highlighted facial emotion recognition (FER) as an essential component of human–computer interaction (HCI). Traditional FER studies based on handcrafted features and shallow machine learning have shown a limited performance, while convolutional neural networks (CNNs) have improved nonlinear emotion pattern analysis but have been constrained by local feature extraction. Vision transformers (ViTs) have addressed this by leveraging global correlations, yet both CNN- and ViT-based single networks often suffer from overfitting, single-network dependency, and information loss in ensemble operations. To overcome these limitations, we propose ArecaNet, an assembled residual enhanced cross-attention network that integrates multiple feature streams without information loss. The framework comprises (i) channel and spatial feature extraction via SCSESResNet, (ii) landmark feature extraction from specialized sub-networks, (iii) iterative fusion through residual enhanced cross-attention, (iv) final emotion classification from the fused representation. Our research introduces a novel approach by integrating pre-trained sub-networks specialized in facial recognition with an attention mechanism and our uniquely designed main network, which is optimized for size reduction and efficient feature extraction. The extracted features are fused through an iterative residual enhanced cross-attention mechanism, which minimizes information loss and preserves complementary representations across networks. This strategy overcomes the limitations of conventional ensemble methods, enabling seamless feature integration and robust recognition. The experimental results show that the proposed ArecaNet achieved accuracies of 97.0% and 97.8% using the public databases, FER-2013 and RAF-DB, which were 4.5% better than the existing state-of-the-art method, PAtt-Lite, for FER-2013 and 2.75% for RAF-DB, and achieved a new state-of-the-art accuracy for each database. Full article
(This article belongs to the Special Issue Sensor-Based Behavioral Biometrics)
►▼ Show Figures

Figure 1

23 pages, 3668 KB  
Article
Analysis and Simulation Verification of the Verticality Measurement Model for Single Offshore Pile Based on Binocular Vision
by Shaohui Li, Yanlong Zhu, Yuanyuan Cao, Xinghua Li and Zhenjie Zhou
Sensors 2025, 25(23), 7374; https://doi.org/10.3390/s25237374 - 4 Dec 2025
Viewed by 382
Abstract
Accurately measuring the verticality of a single pile is of crucial importance for ensuring the safe operation of offshore wind power projects. However, mainstream methods have disadvantages such as high dependence on manual labor, low real-time performance, and susceptibility to construction site conditions [...] Read more.
Accurately measuring the verticality of a single pile is of crucial importance for ensuring the safe operation of offshore wind power projects. However, mainstream methods have disadvantages such as high dependence on manual labor, low real-time performance, and susceptibility to construction site conditions and marine environmental impacts. The method of measuring the verticality of a single offshore pile based on binocular vision is one of the emerging measurement methods, but there is currently a lack of research on measurement models. In order to clarify the principle of the method for measuring the verticality of a single pile at sea based on binocular vision, this paper starts from the imaging principle of the camera and studies and derives the measurement model of the verticality of a single pile in the global coordinate system and the error model of the measurement system. To verify the correctness of the model and method, a testing experimental platform was built to simulate the measurement of the ship under static and dynamic conditions, and the measurement results were compared with those of the total station. The experimental results show that in the static simulation experiment, the maximum absolute error of the verticality of a single pile is 0.2°, the maximum absolute error of the roll angle is 0.3°, and the maximum absolute error of the pitch angle is 0.3°. In the dynamic simulation experiment, the maximum absolute error of the verticality of a single pile is 0.4°, the maximum absolute error of the roll angle is 0.3°, and the maximum absolute error of the pitch angle is 0.3°. This paper verified the correctness of the model and provided model support for measuring the verticality of single piles at sea. Full article
(This article belongs to the Topic Applied Computer Vision and Pattern Recognition: 2nd Edition)
►▼ Show Figures

Figure 1

19 pages, 7350 KB  
Article
Impact Mechanism of Spectral Differentiation on PV Performance and Optimization of PV Systems in Shaded Forest Environments
by Dongxiao Yang, Yuan He, Latai Ga, Daochun Xu, Xiaopeng Bai and Wenbin Li
Sensors 2025, 25(23), 7373; https://doi.org/10.3390/s25237373 - 4 Dec 2025
Viewed by 325
Abstract
The global low-carbon transition is driving the use of renewable energy for ecological monitoring. Traditional power supply for forest monitoring sensor equipment is constrained by high wired costs, frequent battery replacement, and the limitations of low light levels and special spectra under forest [...] Read more.
The global low-carbon transition is driving the use of renewable energy for ecological monitoring. Traditional power supply for forest monitoring sensor equipment is constrained by high wired costs, frequent battery replacement, and the limitations of low light levels and special spectra under forest canopies on photovoltaic (PV) compatibility. Existing research lacks exploration of the correlation between under-forest spectra and PV performance. This study measured the summer understory light spectra of five tree species in Beijing, evaluated the performance of three types of PV cells—monocrystalline silicon, polycrystalline silicon, and amorphous silicon—and designed a low-light energy harvesting circuit. Results indicate that spectral differences under tree canopies are concentrated from 380–680 nm, exhibiting a distinctive forest-specific spectral feature of “high-band enrichment” above 680 nm. Under low-light conditions, polycrystalline silicon photovoltaics demonstrates optimal performance when adapted to this high-band spectrum. The designed circuit can activate at 5 W/m2 irradiance and stably output 4.16 V voltage. This study fills a spectral gap in northern summer tree canopies, providing a comprehensive solution of “material adaptation + circuit customization” for the practical deployment of shaded forest PV systems. Full article
(This article belongs to the Section Optical Sensors)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-302
Previous Issue
Next Issue
Back to TopTop