Previous Issue
Volume 25, October-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 21
sensors-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Sensors, Volume 25, Issue 21 (November-1 2025) – 65 articles

  • 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:
26 pages, 12008 KB  
Article
A Secure and Lightweight ECC-Based Authentication Protocol for Wireless Medical Sensors Networks
by Yu Shang, Junhua Chen, Shenjin Wang, Ya Zhang and Kaixuan Ma
Sensors 2025, 25(21), 6567; https://doi.org/10.3390/s25216567 (registering DOI) - 24 Oct 2025
Abstract
Wireless Medical Sensor Networks (WMSNs) collect and transmit patients’ physiological data in real time through various sensors, playing an increasingly important role in intelligent healthcare. Authentication protocols in WMSNs ensure that users can securely access real-time data from sensor nodes. Although many researchers [...] Read more.
Wireless Medical Sensor Networks (WMSNs) collect and transmit patients’ physiological data in real time through various sensors, playing an increasingly important role in intelligent healthcare. Authentication protocols in WMSNs ensure that users can securely access real-time data from sensor nodes. Although many researchers have proposed authentication schemes to resist common attacks, insufficient attention has been paid to insider attacks and ephemeral secret leakage (ESL) attacks. Moreover, existing adversary models still have limitations in accurately characterizing an attacker’s capabilities. To address these issues, this paper extends the traditional adversary model to better reflect practical deployment scenarios, assuming a semi-trusted server and allowing adversaries to obtain users’ temporary secrets. Based on this enhanced model, we design an efficient ECC-based authentication and key agreement protocol that ensures the confidentiality of users’ passwords, biometric data, and long-term private keys during the registration phase, thereby mitigating insider threats. The proposed protocol combines anonymous authentication and elliptic curve cryptography (ECC) key exchange to satisfy security requirements. Performance analysis demonstrates that the proposed protocol achieves lower computational and communication costs compared with existing schemes. Furthermore, the protocol’s security is formally proven under the Random Oracle (ROR) model and verified using the ProVerif tool, confirming its security and reliability. Therefore, the proposed protocol can be effectively applied to secure data transmission and user authentication in wireless medical sensor networks and other IoT environments. Full article
(This article belongs to the Section Biomedical Sensors)
15 pages, 831 KB  
Article
PM2.5 Pollution Decrease in Paris, France, for the 2013–2024 Period: An Evaluation of the Local Source Contributions by Subtracting the Effect of Wind Speed
by Jean-Baptiste Renard and Jérémy Surcin
Sensors 2025, 25(21), 6566; https://doi.org/10.3390/s25216566 (registering DOI) - 24 Oct 2025
Abstract
Measuring the long-term trend of PM2.5 mass-concentration in urban environments is essential as it has a direct impact on human health. PM2.5 levels depend not only on the intensity of local emission sources and on imported pollution, but also on meteorological conditions (e.g., [...] Read more.
Measuring the long-term trend of PM2.5 mass-concentration in urban environments is essential as it has a direct impact on human health. PM2.5 levels depend not only on the intensity of local emission sources and on imported pollution, but also on meteorological conditions (e.g., anticyclonic versus windy conditions), which leads to yearly variations in mean PM2.5 values. Two datasets available for Paris, France, are considered: measurements from Airparif air quality agency network and from the Pollutrack network of mobile car-based sensors. Also, meteorological parameters coming from ERA5 analysis (ECMWF) are considered. Annual values are calculated using three different statistical methods, which yield different results. For the 2013–2024 period, a clear relationship between wind speed and PM2.5 mass-concentration levels is established. The results show a linear decrease in both concentration and standard deviation for wind speeds in the 0–6 m.s−1 range, followed by nearly stable values for wind speed above 6 m.s−1. This behavior is explained by the dispersive effect of strong winds on air pollution. Under such conditions, which occur about 10% of the time in Paris, the contribution of persistent background sources can be isolated. Using the 6 m·s−1 threshold, the average annual linear decrease in emissions from local sources is estimated at 4.1 and 4.3% per year for the Airparif and Pollutrack data, respectively. Since 2023, the annual background value attributed to emission has been close to 5 µg.m−3, in agreement with WHO recommendations. This approach could be used to monitor the effects of regulations on traffic and heating emissions and could be applied to other cities for estimating background pollution levels. Finally, future studies should therefore prioritize number concentrations and size distributions, rather than mass-concentrations. Full article
(This article belongs to the Section Environmental Sensing)
28 pages, 3439 KB  
Article
The Task Dependency of Spontaneous Rhythmic Performance in Movements Beyond Established Biomechanical Models: An Inertial Sensor-Based Study
by Analina Emmanouil, Fani Paderi, Konstantinos Boudolos and Elissavet Rousanoglou
Sensors 2025, 25(21), 6565; https://doi.org/10.3390/s25216565 (registering DOI) - 24 Oct 2025
Abstract
Spontaneous rhythmic performance is a fundamental feature of human movement, well established in biomechanical models (EBMs) but less understood in complex physical fitness exercises (PFEs). This study examined the task dependency of spontaneous rhythmic performance across three EBMs (walking, hopping, finger tapping) and [...] Read more.
Spontaneous rhythmic performance is a fundamental feature of human movement, well established in biomechanical models (EBMs) but less understood in complex physical fitness exercises (PFEs). This study examined the task dependency of spontaneous rhythmic performance across three EBMs (walking, hopping, finger tapping) and seven PFEs (hip abduction, back extension, sit-up, push-up, shoulder abduction, squat, lunge). A total of 15 men and 15 women performed each task at a self-selected pace while wearing inertial sensors. Measures included spontaneous motor tempo (SMT), temporal structure metrics, and their within- and between-trial individual variability (%CV) (ANOVA, SPSS 28.0, p ≤ 0.05). SMT was task-dependent, with EMB tasks being near ~2 Hz (walking: 1.82 ± 0.10 Hz; hopping: 2.08 ± 0.22 Hz; finger tapping: 1.89 ± 0.43 Hz) and PFEs being slower (0.36–0.68 Hz). Temporal structure mirrored these differences with shorter cycle and phase durations in EBM than PFE tasks, with relative phase durations consistently at about a 1:1 ratio. Τhe overall low %CV indicated stable performance (within-trial: 1.4–7.5%; between-trial: 0.5–7.8%). The results highlight the task dependency of SMT and temporal structure, as well as the robustness of an overarching internal timing framework supporting rhythmic motor control across diverse movement contexts. Full article
(This article belongs to the Special Issue Innovative Applications of Wearable Sensors in Musculoskeletal Biomechanics and Rehabilitation)
24 pages, 12791 KB  
Article
Enabling Efficient Scheduling of Multi-Type Sources in Power Systems via Uncertainty Monitoring and Nonlinear Constraint Processing
by Di Zhang, Qionglin Li, Ji Han, Chunsun Tian and Yebin Li
Sensors 2025, 25(21), 6564; https://doi.org/10.3390/s25216564 (registering DOI) - 24 Oct 2025
Abstract
The large-scale integration of renewable energy sources introduces significant uncertainty into modern power systems, posing new challenges for reliable and economical operation. Effective scheduling therefore requires accurate monitoring of uncertainty and efficient handling of nonlinear system dynamics. This paper proposes an optimization-based scheduling [...] Read more.
The large-scale integration of renewable energy sources introduces significant uncertainty into modern power systems, posing new challenges for reliable and economical operation. Effective scheduling therefore requires accurate monitoring of uncertainty and efficient handling of nonlinear system dynamics. This paper proposes an optimization-based scheduling method that combines sensor-informed monitoring of photovoltaic (PV) uncertainty with advanced processing of nonlinear hydropower characteristics. A detailed hydropower model is incorporated into the framework to represent water balance, reservoir dynamics, and head–discharge–power relationships with improved accuracy. Nonlinear constraints and uncertainty are addressed through a unified approximation scheme that ensures computational tractability. Case studies on the modified IEEE−39 system show that the proposed method achieves effective multi-source coordination, reduces operating costs by up to 2.9%, and enhances renewable energy utilization across different uncertainty levels and PV penetration scenarios. Full article
(This article belongs to the Special Issue Advanced Sensors and Artificial Intelligence for Condition Monitoring of Power Electronic Systems)
15 pages, 1516 KB  
Article
Development of 3D-Stacked 1Megapixel Dual-Time-Gated SPAD Image Sensor with Simultaneous Dual Image Output Architecture for Efficient Sensor Fusion
by Kazuma Chida, Kazuhiro Morimoto, Naoki Isoda, Hiroshi Sekine, Tomoya Sasago, Yu Maehashi, Satoru Mikajiri, Kenzo Tojima, Mahito Shinohara, Ayman T. Abdelghafar, Hiroyuki Tsuchiya, Kazuma Inoue, Satoshi Omodani, Alice Ehara, Junji Iwata, Tetsuya Itano, Yasushi Matsuno, Katsuhito Sakurai and Takeshi Ichikawa
Sensors 2025, 25(21), 6563; https://doi.org/10.3390/s25216563 (registering DOI) - 24 Oct 2025
Abstract
Sensor fusion is crucial in numerous imaging and sensing applications. Integrating data from multiple sensors with different field-of-view, resolution, and frame timing poses substantial computational overhead. Time-gated single-photon avalanche diode (SPAD) image sensors have been developed to support multiple sensing modalities and mitigate [...] Read more.
Sensor fusion is crucial in numerous imaging and sensing applications. Integrating data from multiple sensors with different field-of-view, resolution, and frame timing poses substantial computational overhead. Time-gated single-photon avalanche diode (SPAD) image sensors have been developed to support multiple sensing modalities and mitigate this issue, but mismatched frame timing remains a challenge. Dual-time-gated SPAD image sensors, which can capture dual images simultaneously, have also been developed. However, the reported sensors suffered from medium-to-large pixel pitch, limited resolution, and inability to independently control the exposure time of the dual images, which restricts their applicability. In this paper, we introduce a 5 µm-pitch, 3D-backside-illuminated (BSI) 1Megapixel dual-time-gated SPAD image sensor enabling a simultaneous output of dual images. The developed SPAD image sensor is verified to operate as an RGB-Depth (RGB-D) sensor without complex image alignment. In addition, a novel high dynamic range (HDR) technique, utilizing pileup effect with two parallel in-pixel memories, is validated for dynamic range extension in 2D imaging, achieving a dynamic range of 119.5 dB. The proposed architecture provides dual image output with the same field-of-view, resolution, and frame timing, and is promising for efficient sensor fusion. Full article
(This article belongs to the Special Issue Special Issue on the 2025 International Image Sensor Workshop (IISW2025))
39 pages, 3305 KB  
Article
A Robust and Efficient Workflow for Heart Valve Disease Detection from PCG Signals: Integrating WCNN, MFCC Optimization, and Signal Quality Evaluation
by Shin-Chi Lai, Yen-Ching Chang, Ying-Hsiu Hung, Szu-Ting Wang, Yao-Feng Liang, Li-Chuan Hsu, Ming-Hwa Sheu and Chuan-Yu Chang
Sensors 2025, 25(21), 6562; https://doi.org/10.3390/s25216562 (registering DOI) - 24 Oct 2025
Abstract
This study proposes a comprehensive and computationally efficient system for the recognition of heart valve diseases (HVDs) in phonocardiogram (PCG) signals, emphasizing an end-to-end workflow suitable for real-world deployment. The core of the system is a lightweight weighted convolutional neural network (WCNN) featuring [...] Read more.
This study proposes a comprehensive and computationally efficient system for the recognition of heart valve diseases (HVDs) in phonocardiogram (PCG) signals, emphasizing an end-to-end workflow suitable for real-world deployment. The core of the system is a lightweight weighted convolutional neural network (WCNN) featuring a key weighting calculation (KWC) layer, which enhances noise robustness by adaptively weighting feature map channels based on global average pooling. The proposed system incorporates optimized feature extraction using Mel-frequency cepstral coefficients (MFCCs) guided by GradCAM, and a band energy ratio (BER) metric to assess signal quality, showing that lower BER values are associated with higher misclassification rates due to noise. Experimental results demonstrated classification accuracies of 99.6% and 90.74% on the GitHub PCG and PhysioNet/CinC Challenge 2016 databases, respectively, where the models were trained and tested independently. The proposed model achieved superior accuracy using significantly fewer parameters (312,357) and lower computational cost (4.5 M FLOPs) compared with previously published research. Compared with the model proposed by Karhade et al., the proposed model use 74.9% fewer parameters and 99.3% fewer FLOPs. Furthermore, the proposed model was implemented on a Raspberry Pi, achieving real-time HVDs detection with a detection time of only 1.87 ms for a 1.4 s signal. Full article
(This article belongs to the Special Issue AI-Based Automated Recognition and Detection in Healthcare)
24 pages, 3478 KB  
Article
Measurement of Force and Position Using a Cantilever Beam and Multiple Strain Gauges: Sensing Principles and Design Considerations
by Carter T. Noh, Kenneth Smith, Christian L. Shamo, Jordan Porter, Kirsten Steele, Nathan D. Ludlow, Ryan W. Hall, Maeson G. Holst, Alex R. Williams and Douglas D. Cook
Sensors 2025, 25(21), 6561; https://doi.org/10.3390/s25216561 (registering DOI) - 24 Oct 2025
Abstract
Simultaneous measurement of force and position often relies on delicate tactile sensing systems that only measure small forces at discrete positions. This study proposes a compact, durable sensor which can provide simultaneous and continuous measurements of force and position using multiple strain gauges [...] Read more.
Simultaneous measurement of force and position often relies on delicate tactile sensing systems that only measure small forces at discrete positions. This study proposes a compact, durable sensor which can provide simultaneous and continuous measurements of force and position using multiple strain gauges mounted on a cantilever beam. When a point force is applied to the cantilever, the strain gauges are used to determine the magnitude of the applied force and its position along the beam. A major advantage of the force-position sensor concept is its compact electronics and durable sensing surface. We designed, tested, and evaluated three different prototypes for the force-position sensor concept. The prototypes achieved an average percent error of 1.71% and were highly linear. We also conducted a thorough analysis of design variables and their effects on performance. The force and position measurement ranges can be adjusted by tuning the material and geometric properties of the beam and the spacing of the strain gauges. The accuracy of force measurements is dependent upon applied load, but insensitive to the location of the applied load. Accuracy of position measurements is also dependent upon applied load and weakly dependent upon position of the applied load. Full article
(This article belongs to the Collection Tactile Sensors, Sensing and Systems)
►▼ Show Figures

Figure 1

26 pages, 1644 KB  
Article
Context-Aware Alerting in Elderly Care Facilities: A Hybrid Framework Integrating LLM Reasoning with Rule-Based Logic
by Nazmun Nahid, Md Atiqur Rahman Ahad and Sozo Inoue
Sensors 2025, 25(21), 6560; https://doi.org/10.3390/s25216560 (registering DOI) - 24 Oct 2025
Abstract
The rising demand for elderly care amid ongoing nursing shortages has highlighted the limitations of conventional alert systems, which frequently generate excessive alerts and contribute to alarm fatigue. The objective of this study is to develop a hybrid, context-aware nurse alerting framework for [...] Read more.
The rising demand for elderly care amid ongoing nursing shortages has highlighted the limitations of conventional alert systems, which frequently generate excessive alerts and contribute to alarm fatigue. The objective of this study is to develop a hybrid, context-aware nurse alerting framework for long-term care (LTC) facilities that minimizes redundant alarms, reduces alarm fatigue, and enhances patient safety and caregiving balance during multi-person care scenarios such as mealtimes. To do so, we aimed to intelligently suppress, delay, and validate alerts by integrating rule-based logic with Large Language Model (LLM)-driven semantic reasoning. We conducted an experimental study in a real-world LTC environment involving 28 elderly residents (6 high, 8 medium, and 14 low care levels) and four nurses across three rooms over seven days. The proposed system utilizes video-derived skeletal motion, care-level annotations, and dynamic nurse–elderly proximity for decision making. Statistical analyses were performed using F1 score, accuracy, false positive rate (FPR), and false negative rate (FNR) to evaluate performance improvements. Compared to the baseline where all nurses were notified (100% alarm load), the proposed method reduced average alarm load to 27.5%, achieving a 72.5% reduction, with suppression rates reaching 100% in some rooms for some nurses. Performance metrics further validate the system’s effectiveness: the macro F1 score improved from 0.18 (baseline) to 0.97, while accuracy rose from 0.21 (baseline) to 0.98. Compared to the baseline error rates (FPR 0.20, FNR 0.79), the proposed method achieved drastically lower values (FPR 0.005, FNR 0.023). Across both spatial (room-level) and temporal (day-level) validations, the proposed approach consistently outperformed baseline and purely rule-based methods. These findings demonstrate that the proposed approach effectively minimizes false alarms while maintaining strong operational efficiency. By integrating rule-based mechanisms with LLM-based contextual reasoning, the framework significantly enhances alert accuracy, mitigates alarm fatigue, and promotes safer, more sustainable, and human-centered care practices, making it suitable for practical deployment within real-world long-term care environments. Full article
(This article belongs to the Section Biomedical Sensors)
►▼ Show Figures

Graphical abstract

15 pages, 1823 KB  
Article
Improved Quadtree-Based Selection of Single Images for 3D Generation
by Wanyun Li, Yue Liu, Yuqiang Fang, Yasheng Zhang, Yao Lu and Gege Sun
Sensors 2025, 25(21), 6559; https://doi.org/10.3390/s25216559 (registering DOI) - 24 Oct 2025
Abstract
With the rapid development of large generative models for 3D content, image-to-3D and text-to-3D generation has become a major focus in computer vision and graphics. Single-view 3D reconstruction, in particular, offers a convenient and practical solution. However, the way to automatically choose the [...] Read more.
With the rapid development of large generative models for 3D content, image-to-3D and text-to-3D generation has become a major focus in computer vision and graphics. Single-view 3D reconstruction, in particular, offers a convenient and practical solution. However, the way to automatically choose the best image from a large collection to optimize reconstruction quality and efficiency is very important. This paper proposes a novel image selection framework based on multi-feature fusion quadtree structure. Here, we introduce a new image selection method based on a multi-feature quadtree structure. Our approach integrates various visual and semantic features and uses a hierarchical quadtree to efficiently evaluate image content. This allows us to identify the most informative and reconstruction-friendly image from large datasets. We then use Tencent’s Hunyuan 3D model to verify that the selected image improves reconstruction performance. Experimental results show that our method outperforms existing approaches across key metrics. Baseline methods achieved average scores of 6.357 in Accuracy, 6.967 in Completeness, and 6.662 Overall. Our method reduced these to 4.238, 5.166, and 4.702, corresponding to an average error reduction of 29.5%. These results confirm that our approach reduces reconstruction errors, improves geometric consistency, and yields more visually plausible 3D models. Full article
(This article belongs to the Section Sensing and Imaging)
►▼ Show Figures

Figure 1

15 pages, 791 KB  
Article
Effects of Sanda Sports Training on Cognitive–Motor Control Based on EEG and Heart Rate Sensors: A Coupled ERP and HRV Analysis
by Ziwen Ning, Jiayi Zhao, Chuanyin Jiang, Haojie Li, Haidong Jiang and Tianfen Zhou
Sensors 2025, 25(21), 6558; https://doi.org/10.3390/s25216558 (registering DOI) - 24 Oct 2025
Abstract
Objective: To investigate whether prolonged Sanda combat experience improves cognitive–motor control via neuro-cardiac coupling. Methods: Nineteen national-level Sanda athletes and nineteen matched controls completed a color-word Stroop task while concurrent EEG and ECG were recorded. The conflict adaptation effect (CAE), which [...] Read more.
Objective: To investigate whether prolonged Sanda combat experience improves cognitive–motor control via neuro-cardiac coupling. Methods: Nineteen national-level Sanda athletes and nineteen matched controls completed a color-word Stroop task while concurrent EEG and ECG were recorded. The conflict adaptation effect (CAE), which refers to the ability to adjust cognitive control in response to conflicting stimuli, was compared between groups, along with P600 and LSP amplitudes and heart rate variability (RMSSD, HF); mediation analysis examined vagal recovery as a pathway. Results: Athletes responded faster and showed a larger CAE than controls (p < 0.001). ERP analyses revealed larger CAE-related P600 and LSP amplitudes in athletes (p < 0.05), with LSP amplitude inversely correlating with behavioral CAE (p < 0.05). Post-task vagal rebound (ΔRMSSD and ΔHF) was significantly greater in athletes (p < 0.05), and ΔRMSSD positively correlated with CAE (p < 0.05). Mediation analysis confirmed that vagal recovery partially mediated the association between Sanda experience and improved cognitive–motor control (p < 0.05). Conclusions: Sanda training enhances cognitive–motor control by accelerating parasympathetic recovery and optimizing neural conflict processing, providing evidence for an integrated exercise–cognition–autonomic nervous system coupling model. Full article
(This article belongs to the Special Issue Wearable and Portable Devices for Endurance Sports)
23 pages, 26041 KB  
Article
A Portable Measurement System Based on Nanomembranes for Pollutant Detection in Water
by Luca Tari, Maria Cojocari, Gabriele Cavaliere, Sarah Sibilia, Francesco Siconolfi, Georgy Fedorov, Luigi Ferrigno, Polina Kuzhir and Antonio Maffucci
Sensors 2025, 25(21), 6557; https://doi.org/10.3390/s25216557 (registering DOI) - 24 Oct 2025
Abstract
This work presents the design, the development and the experimental validation of a portable, low-cost sensing system for the detection of waterborne pollutants. The proposed system is based on Electrochemical Impedance Spectroscopy and PPF+Ni nanomembrane sensors. Designed in response to the increasing demand [...] Read more.
This work presents the design, the development and the experimental validation of a portable, low-cost sensing system for the detection of waterborne pollutants. The proposed system is based on Electrochemical Impedance Spectroscopy and PPF+Ni nanomembrane sensors. Designed in response to the increasing demand for in situ water quality monitoring, the system integrates a simplified, scalable EIS acquisition architecture compatible with microcontroller-based platforms. The sensing configuration utilises the voltage divider principle, ensuring simplicity in signal conditioning by allowing compatibility with different electrode types through passive impedance matching. In addition, new merit figures have been proposed and implemented to analyse the measures. The proposed platform was experimentally characterised for its measurement stability, accuracy and environmental robustness. Sensitivity tests using benzoquinone as a target analyte demonstrated the capability of detecting concentrations as low as 0.1 mM with a monotonic response over increasing concentrations. A comparative study with a commercial electrochemical system (PalmSens4) under identical conditions highlighted the higher resolution and practical advantages of the proposed method despite operating with a lower impedance range. Additionally, the system exhibited reliable discrimination across tested concentrations and greater adaptability for integration into field-deployable environmental monitoring platforms. Future developments will focus on optimising selectivity through new sensor materials and analytical modelling of uncertainty propagation in the analysis based on defined figures of merit. Full article
(This article belongs to the Special Issue Sensors for Water Quality Monitoring and Assessment)
39 pages, 29667 KB  
Article
Frugal Self-Optimization Mechanisms for Edge–Cloud Continuum
by Zofia Wrona, Katarzyna Wasielewska-Michniewska, Maria Ganzha, Marcin Paprzycki and Yutaka Watanobe
Sensors 2025, 25(21), 6556; https://doi.org/10.3390/s25216556 (registering DOI) - 24 Oct 2025
Abstract
The increasing complexity of the Edge–Cloud Continuum (ECC), driven by the rapid expansion of the Internet of Things (IoT) and data-intensive applications, necessitates implementing innovative methods for automated and efficient system management. In this context, recent studies focused on the utilization of self-* [...] Read more.
The increasing complexity of the Edge–Cloud Continuum (ECC), driven by the rapid expansion of the Internet of Things (IoT) and data-intensive applications, necessitates implementing innovative methods for automated and efficient system management. In this context, recent studies focused on the utilization of self-* capabilities that can be used to enhance system autonomy and increase operational proactiveness. Separately, anomaly detection and adaptive sampling techniques have been explored to optimize data transmission and improve systems’ reliability. The integration of those techniques within a single, lightweight, and extendable self-optimization module is the main subject of this contribution. The module was designed to be well suited for distributed systems, composed of highly resource-constrained operational devices (e.g., wearable health monitors, IoT sensors in vehicles, etc.), where it can be utilized to self-adjust data monitoring and enhance the resilience of critical processes. The focus is put on the implementation of two core mechanisms, derived from the current state-of-the-art: (1) density-based anomaly detection in real-time resource utilization data streams, and (2) a dynamic adaptive sampling technique, which employs Probabilistic Exponential Weighted Moving Average. The performance of the proposed module was validated using both synthetic and real-world datasets, which included a sample collected from the target infrastructure. The main goal of the experiments was to showcase the effectiveness of the implemented techniques in different, close to real-life scenarios. Moreover, the results of the performed experiments were compared with other state-of-the-art algorithms in order to examine their advantages and inherent limitations. With the emphasis put on frugality and real-time operation, this contribution offers a novel perspective on resource-aware autonomic optimization for next-generation ECC. Full article
(This article belongs to the Special Issue Artificial Intelligence and Edge Computing in IoT-Based Applications)
►▼ Show Figures

Figure 1

13 pages, 2511 KB  
Article
NLOS Identification and Error Compensation Method for UWB in Workshop Scene
by Yu Su, Quan Yu, Xiaohao Xia, Wenfeng Li, Lijun He and Taiwei Yang
Sensors 2025, 25(21), 6555; https://doi.org/10.3390/s25216555 (registering DOI) - 24 Oct 2025
Abstract
To address the frequent safety incidents caused by positioning uncertainty due to NLOS (Non-Line-of-Sight) interference in complex manufacturing workshop environments, this paper aims to achieve high-precision distance measurement and positioning in complex workshop scenarios. First, common NLOS identification methods are analyzed. By combining [...] Read more.
To address the frequent safety incidents caused by positioning uncertainty due to NLOS (Non-Line-of-Sight) interference in complex manufacturing workshop environments, this paper aims to achieve high-precision distance measurement and positioning in complex workshop scenarios. First, common NLOS identification methods are analyzed. By combining received signal energy and ranging residuals, a rapid NLOS identification method is proposed. Building on this foundation, a ranging error compensation method based on maximum likelihood estimation and adaptive extended Kalman filtering is designed. Finally, static experiments are conducted to verify the effectiveness of the proposed NLOS identification method and ranging error compensation approach. Experimental results indicate that the ranging accuracy of the proposed method has been significantly improved and demonstrates considerable advantages over traditional Kalman filtering algorithms. Full article
(This article belongs to the Section Industrial Sensors)
►▼ Show Figures

Figure 1

35 pages, 3368 KB  
Article
A Resilient Distributed Pareto-Based PSO for Edge-UAVs Deployment Optimization in Internet of Flying Things
by Sabrina Zerrougui, Sofiane Zaidi and Carlos T. Calafate
Sensors 2025, 25(21), 6554; https://doi.org/10.3390/s25216554 (registering DOI) - 24 Oct 2025
Abstract
Particle Swarm Optimization (PSO) has been widely employed to optimize the deployment of Unmanned Aerial Vehicles (UAVs) in various scenarios, particularly because of its efficiency in handling both single and multi-objective optimization problems. In this paper, a framework for optimizing the deployment of [...] Read more.
Particle Swarm Optimization (PSO) has been widely employed to optimize the deployment of Unmanned Aerial Vehicles (UAVs) in various scenarios, particularly because of its efficiency in handling both single and multi-objective optimization problems. In this paper, a framework for optimizing the deployment of edge-enabled UAVs using Pareto-PSO is proposed for data collection scenarios in which UAVs operate autonomously and execute onboard distributed multi-objective PSO to maximize the total non-overlapping coverage area while minimizing latency and energy consumption. Performance evaluation is conducted using key indicators, including convergence time, throughput, and total non-overlapping coverage area across bandwidth and swarm-size sweeps. Simulation results demonstrate that the Pareto-PSO consistently attains the highest throughput and the largest coverage envelope, while exhibiting moderate and scalable convergence times. These results highlight the advantage of treating the objectives as a vector-valued objective in Pareto-PSO for real-time, scalable, and energy-aware edge-UAV deployment in dynamic Internet of Flying Things environments. Full article
(This article belongs to the Special Issue Wireless Communication Technologies for Internet of Things and Wireless Sensor Networks)
69 pages, 12722 KB  
Review
A Review of Graphene-Integrated Biosensors for Non-Invasive Biochemical Monitoring in Health Applications
by Sourabhi Debnath, Tanmoy Debnath and Manoranjan Paul
Sensors 2025, 25(21), 6553; https://doi.org/10.3390/s25216553 (registering DOI) - 24 Oct 2025
Abstract
This review explores the transformative potential of graphene-based, non-invasive biochemical sensors in the context of real-time health monitoring and personalised medicine. Traditional diagnostic methods often involve invasive procedures that can be uncomfortable, pose risks, and limit the frequency of monitoring. In contrast, wearable [...] Read more.
This review explores the transformative potential of graphene-based, non-invasive biochemical sensors in the context of real-time health monitoring and personalised medicine. Traditional diagnostic methods often involve invasive procedures that can be uncomfortable, pose risks, and limit the frequency of monitoring. In contrast, wearable sensors incorporating graphene offer a compelling alternative by enabling continuous, real-time tracking of physiological and biochemical signals with minimal intrusion. Graphene’s exceptional electrical conductivity, mechanical flexibility, biocompatibility, and high surface-area-to-volume ratio make it ideally suited for integration into skin-conformal sensor platforms. These properties not only enhance sensitivity and signal fidelity but also promote user comfort and long-term wearability, critical factors for the adoption of wearable health technologies. The discussion evaluates current developments in the design and deployment of graphene-based biosensors, with particular attention given to their role in managing chronic conditions, supporting preventative healthcare, and facilitating decentralised diagnostics. By bridging materials science and biomedical engineering, this review positions graphene as a key enabler in the shift towards more proactive, patient-centred healthcare models. The text also identifies ongoing challenges and future directions in sensor design, aiming to inform researchers working at the intersection of advanced materials and medical technology. Full article
(This article belongs to the Section Biomedical Sensors)
17 pages, 11106 KB  
Article
Assessment of Neutron Radiation Effects on the Fiber Optics Current Sensor Performance During JET DTE2 Experimental Campaign
by Andrei Gusarov, Perry Beaumont and JET Contributors
Sensors 2025, 25(21), 6552; https://doi.org/10.3390/s25216552 (registering DOI) - 24 Oct 2025
Abstract
Fibre Optics Current Sensor (FOCS) will be used at ITER to perform plasma current measurement during quasi-steady state D-T plasma operation. Effects of the tokamak harsh environment on the FOCS performance must be evaluated to predict possible failure modes and relevant mitigation measures. [...] Read more.
Fibre Optics Current Sensor (FOCS) will be used at ITER to perform plasma current measurement during quasi-steady state D-T plasma operation. Effects of the tokamak harsh environment on the FOCS performance must be evaluated to predict possible failure modes and relevant mitigation measures. The influence of nuclear radiation with the significant flux of 14 MeV neutrons is of specific concern. This problem was addressed by operating the FOCS during D-T campaign at JET (DTE2). In the present report experimental results are presented and analysed. These results indicate that FOCS will effectively perform current measurements during ITER nuclear operation. Full article
(This article belongs to the Special Issue Optical Sensors for Industrial Applications)
►▼ Show Figures

Figure 1

18 pages, 1921 KB  
Article
Validity of a Convolutional Neural Network-Based, Markerless Pose Estimation System Compared to a Marker-Based 3D Motion Analysis System for Gait Assessment—A Pilot Study
by Korbinian Ksoll, Rafael Krätschmer and Fabian Stöcker
Sensors 2025, 25(21), 6551; https://doi.org/10.3390/s25216551 (registering DOI) - 24 Oct 2025
Abstract
Gait analysis is a valuable tool for a wide range of clinical applications. Until now, the standard for gait analysis has been marker-based 3D optical systems. Recently, markerless gait analysis systems that utilize pose estimation models based on Convolutional Neural Networks (CNNs) and [...] Read more.
Gait analysis is a valuable tool for a wide range of clinical applications. Until now, the standard for gait analysis has been marker-based 3D optical systems. Recently, markerless gait analysis systems that utilize pose estimation models based on Convolutional Neural Networks (CNNs) and computer vision have gained importance. In this pilot study, we validated the performance of a CNN-based, markerless pose estimation algorithm (Orthelligent® VISION; OV) compared to a standard marker-based 3D motion capture system in 16 healthy adults. Standard gait metrics were analyzed by calculating concordance correlation coefficients (CCCs) and coefficients of variation. With regard to gait event detection, we found good overlaps for both systems. Compared to the marker-based motion analysis, OV achieved a strong to almost complete concordance regarding the sagittal measurement of cadence, gait variability, step time, stance time, step length, and double support (CCC ≥ 0.624), as well as regarding the frontal plane parameters of cadence, step time, stance time, and step width (CCC ≥ 0.805). For gait symmetry only, we found a moderate to weak correlation. These results support the CNN-based, markerless gait analysis system OV as an alternative to marker-based 3D motion capture systems for a broad variety of clinical applications. Full article
(This article belongs to the Special Issue Advanced Artificial Devices and Sensing Technologies in Rehabilitation)
►▼ Show Figures

Figure 1

14 pages, 4834 KB  
Article
Crowd Gathering Detection Method Based on Multi-Scale Feature Fusion and Convolutional Attention
by Kamil Yasen, Juting Zhou, Nan Zhou, Ke Qin, Zhiguo Wang and Ye Li
Sensors 2025, 25(21), 6550; https://doi.org/10.3390/s25216550 (registering DOI) - 24 Oct 2025
Abstract
With rapid urbanization and growing population inflows into metropolitan areas, crowd gatherings have become increasingly frequent and dense, posing significant challenges to public safety management. Although existing crowd gathering detection methods have achieved notable progress, they still face major limitations: most rely heavily [...] Read more.
With rapid urbanization and growing population inflows into metropolitan areas, crowd gatherings have become increasingly frequent and dense, posing significant challenges to public safety management. Although existing crowd gathering detection methods have achieved notable progress, they still face major limitations: most rely heavily on local texture or density features and lack the capacity to model contextual information, making them ineffective under severe occlusions and complex backgrounds. Additionally, fixed-scale feature extraction strategies struggle to adapt to crowd regions with varying densities and scales, and insufficient attention to densely populated areas hinders the capture of critical local features. To overcome these challenges, we propose a point-supervised framework named Multi-Scale Convolutional Attention Network (MSCANet). MSCANet adopts a context-aware architecture and integrates multi-scale feature extraction modules and convolutional attention mechanisms, enabling it to dynamically adapt to varying crowd densities while focusing on key regions. This enhances feature representation in complex scenes and improves detection performance. Extensive experiments on public datasets demonstrate that MSCANet achieves high counting accuracy and robustness, particularly in dense and occluded environments, showing strong potential for real-world deployment. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

16 pages, 3350 KB  
Article
A Novel Demographic Indicator Fusion Network (DIFNet) for Dynamic Fusion of EEG and Demographic Indicators for Robust Depression Detection
by Chaoliang Wang, Qingshu Zhou, Mengfan Li, Jiaxin Li and Jing Zhao
Sensors 2025, 25(21), 6549; https://doi.org/10.3390/s25216549 (registering DOI) - 24 Oct 2025
Abstract
Electroencephalography (EEG) has proven to be effective for detecting major depressive disorder (MDD), with deep learning models further advancing its potential. However, the performance of these models may be limited by their neglect of demographic factors (e.g., age, sex, and education), which are [...] Read more.
Electroencephalography (EEG) has proven to be effective for detecting major depressive disorder (MDD), with deep learning models further advancing its potential. However, the performance of these models may be limited by their neglect of demographic factors (e.g., age, sex, and education), which are known to influence EEG characteristics of depression. To address this, we propose DIFNet, a deep learning framework that dynamically fuses EEG features with demographic indicators (age, sex, and years of education) to enhance depression recognition accuracy. DIFNet is composed of four modules: a multiscale convolutional module, a Transformer encoder module, a temporal convolutional network (TCN) module, and a demographic indicator fusion module. The fusion model leverages convolution to process demographic vectors and integrates them with spatiotemporal EEG features, thereby embedding demographic indicators within the deep learning model for classification. Cross-validation between data trials showed that the DIFNet fusing age and years of education achieves a superior accuracy of 99.66%; the dynamic fusion mechanism improves accuracy by 0.72% compared to the baseline without fusing demographic indicators (98.94%), outperforming state-of-the-art methods (SparNet 94.37% and DBGCN 98.30%). Full article
(This article belongs to the Collection EEG-Based Brain–Computer Interface for a Real-Life Appliance)
►▼ Show Figures

Figure 1

2 pages, 693 KB  
Correction
Correction: Meyer zu Westerhausen et al. Optimisation of Sensor and Sensor Node Positions for Shape Sensing with a Wireless Sensor Network—A Case Study Using the Modal Method and a Physics-Informed Neural Network. Sensors 2025, 25, 5573
by Sören Meyer zu Westerhausen, Imed Hichri, Kevin Herrmann and Roland Lachmayer
Sensors 2025, 25(21), 6548; https://doi.org/10.3390/s25216548 (registering DOI) - 24 Oct 2025
Abstract
Mistakes occurred in the original publication [...] Full article
(This article belongs to the Special Issue Sensing and Data Elaboration in Structural Health Monitoring Technology)
►▼ Show Figures

Figure 13

23 pages, 5345 KB  
Article
Vibration Analysis of Aviation Electric Propulsion Test Stand with Active Main Rotor
by Rafał Kliza, Mirosław Wendeker, Paweł Drozd and Ksenia Siadkowska
Sensors 2025, 25(21), 6547; https://doi.org/10.3390/s25216547 (registering DOI) - 24 Oct 2025
Abstract
This paper focuses on the vibration analysis of a prototype helicopter rotor test stand, with particular attention to the dynamic response of its electric propulsion system. The stand is driven by an induction motor and equipped with composite rotor blades of various geometries, [...] Read more.
This paper focuses on the vibration analysis of a prototype helicopter rotor test stand, with particular attention to the dynamic response of its electric propulsion system. The stand is driven by an induction motor and equipped with composite rotor blades of various geometries, including blades with shape memory alloy (SMA)-based torsion actuators for angle of attack (AoA) adjustment. These variable geometries significantly influence the system’s dynamic behavior, where resonance phenomena may pose risks to structural integrity. The objective was to investigate how selected operational parameters specifically motor speed and AoA affect the vibration response of the propulsion system. Structural vibrations were measured using a tri-axial piezoelectric accelerometer system integrated with calibrated signal conditioning and high-resolution data acquisition modules. This setup enabled precise, time-synchronized recording of dynamic responses along all three axes. Fast Fourier Transform (FFT) and Power Spectral Density (PSD) methods were applied to identify dominant frequency components, including those associated with rotor harmonics and SMA activation. The highest vibration amplitudes were observed at an AoA of 16°, but all results remained within the vibration limits defined by MIL-STD-810H for rotorcraft drive systems. The study confirms the importance of sensor-based diagnostics in evaluating electromechanical propulsion systems operating under dynamic loading conditions. Full article
(This article belongs to the Section Fault Diagnosis & Sensors)
►▼ Show Figures

Figure 1

27 pages, 6278 KB  
Article
Evaluation of the Mechanical Stability of Optical Payloads for Remote Sensing Satellites Based on Analysis and Testing Results
by Dulat Akzhigitov, Berik Zhumazhanov, Aigul Kulakayeva, Beksultan Zhumazhanov and Alikhan Kapar
Sensors 2025, 25(21), 6546; https://doi.org/10.3390/s25216546 - 24 Oct 2025
Abstract
This paper presents the results of numerical modeling and vibration testing of a nanosatellite’s optical payload, aimed at assessing its mechanical stability under the mechanical impacts of launch. The purpose of the study is to compare finite element modeling (FEM) data with experimental [...] Read more.
This paper presents the results of numerical modeling and vibration testing of a nanosatellite’s optical payload, aimed at assessing its mechanical stability under the mechanical impacts of launch. The purpose of the study is to compare finite element modeling (FEM) data with experimental testing to refine the computational model and improve the reliability of mechanical stability predictions. The methodology included an FEM analysis with an average damping coefficient, an adapter blank test, a resonance study with a low-level sinusoidal run, random vibration tests, and a sinusoidal pulse test. The FEM results showed an average yield margin of safety MoS = 2.5 with a minimum MoS = 1.8 in the primary mirror mount area. The adapter blank test confirmed the absence of natural resonances in the operating range. The resonance study revealed modes in the 300–1340 Hz range, with the most pronounced peaks in the secondary mirror bracket (520–600 Hz) and the electronics unit (1030–1100 Hz). A comparison of the root mean square (RMS) acceleration values between calculations and tests revealed discrepancies due to the heterogeneous nature of the damping. The values of ζ determined by the half-power method varied from 0.9% to 4.8%, which confirms the dependence of the damping properties on the frequency and localization of the modes. The obtained results confirmed the structural integrity of the payload, allowed for the localization of structural elements, and substantiated the need to consider actual damping coefficients in FEM models. The presented data can be used to optimize the design and improve mechanical stability during payload integration into the satellite platform. Full article
(This article belongs to the Section Sensing and Imaging)
►▼ Show Figures

Figure 1

18 pages, 2578 KB  
Article
Emotion Recognition Using Temporal Facial Skin Temperature and Eye-Opening Degree During Digital Content Viewing for Japanese Older Adults
by Rio Tanabe, Ryota Kikuchi, Min Zou, Kenji Suehiro, Nobuaki Takahashi, Hiroki Saito, Takuya Kobayashi, Hisami Satake, Naoko Sato and Yoichi Kageyama
Sensors 2025, 25(21), 6545; https://doi.org/10.3390/s25216545 - 24 Oct 2025
Abstract
Electroencephalography is a widely used method for emotion recognition. However, it requires specialized equipment, leading to high costs. Additionally, attaching devices to the body during such procedures may cause physical and psychological stress to participants. These issues are addressed in this study by [...] Read more.
Electroencephalography is a widely used method for emotion recognition. However, it requires specialized equipment, leading to high costs. Additionally, attaching devices to the body during such procedures may cause physical and psychological stress to participants. These issues are addressed in this study by focusing on physiological signals that are noninvasive and contact-free, and a generalized method for estimating emotions is developed. Specifically, the facial skin temperature and eye-opening degree of participants captured via infrared thermography and visible cameras are utilized, and emotional states are estimated while Japanese older adults view digital content. Emotional responses while viewing digital content are often subtle and dynamic. Additionally, various emotions occur during such situations, both positive and negative. Fluctuations in facial skin temperature and eye-opening degree reflect activities in the autonomic nervous system. In particular, expressing emotions through facial expressions is difficult for older adults; as such, emotional estimation using such ecological information is required. Our study results demonstrated that focusing on skin temperature changes and eye movements during emotional arousal and non-arousal using bidirectional long short-term memory yields an F1 score of 92.21%. The findings of this study can enhance emotion recognition in digital content, improving user experience and the evaluation of digital content. Full article
(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health: 2nd Edition)
►▼ Show Figures

Figure 1

29 pages, 2298 KB  
Article
Artificial Intelligence and Circadian Thresholds for Stress Detection in Dairy Cattle
by Samuel Lascano Rivera, Luis Rivera, Hernán Benavides and Yasmany Fernández
Sensors 2025, 25(21), 6544; https://doi.org/10.3390/s25216544 - 24 Oct 2025
Abstract
This study investigates stress detection in dairy cattle by integrating circadian rhythm analysis and deep learning. Behavioral biomarkers, including feeding, resting, and rumination, were continuously monitored using Nedap CowControl sensors over a 12-month period to capture seasonal variability. Circadian features were extracted using [...] Read more.
This study investigates stress detection in dairy cattle by integrating circadian rhythm analysis and deep learning. Behavioral biomarkers, including feeding, resting, and rumination, were continuously monitored using Nedap CowControl sensors over a 12-month period to capture seasonal variability. Circadian features were extracted using the Fast Fourier Transform (FFT), and deviations from expected 24 h patterns were quantified using Euclidean distance. These features were used to train a Long Short-Term Memory (LSTM) neural network to classify stress into three levels: normal, mild, and high. Expert veterinary observations of anomalous behaviors and environmental records were used to validate stress labeling. We continuously monitored 10 lactating Holstein cows for 365 days, yielding 87,600 raw hours and 3650 cow-days (one day per cow as the analytical unit). The Short-Time Fourier Transform (STFT, 36 h window, 1 h step) was used solely to derive daily circadian characteristics (amplitude, phase, coherence); STFT windows are not statistical samples. A 60 min window prior to stress onset was incorporated to anticipate stress conditions triggered by management practices and environmental stressors, such as vaccination, animal handling, and cold stress. The proposed LSTM model achieved an accuracy of 82.3% and an AUC of 0.847, outperforming a benchmark logistic regression model (65% accuracy). This predictive capability, with a one-hour lead time, provides a critical window for preventive interventions and represents a practical tool for precision livestock farming and animal welfare monitoring. Full article
(This article belongs to the Special Issue Sensor-Based Behavioral Biometrics)
►▼ Show Figures

Figure 1

22 pages, 4041 KB  
Article
Machine Learning-Based Image Reconstruction in Wearable CC-EIT of the Thorax: Robustness to Electrode Displacement
by Jan Jeschke, Mikhail Ivanenko, Waldemar T. Smolik, Damian Wanta, Mateusz Midura and Przemysław Wróblewski
Sensors 2025, 25(21), 6543; https://doi.org/10.3390/s25216543 - 23 Oct 2025
Abstract
This study investigates the influence of variable electrode positions on image reconstruction in capacitively coupled electrical impedance tomography (CC-EIT) of the human thorax. Images were reconstructed by an adversarial neural network trained on a synthetic dataset generated using a tomographic model that included [...] Read more.
This study investigates the influence of variable electrode positions on image reconstruction in capacitively coupled electrical impedance tomography (CC-EIT) of the human thorax. Images were reconstructed by an adversarial neural network trained on a synthetic dataset generated using a tomographic model that included a wearable elastic band with 32 electrodes attached. Dataset generation was conducted using a previously developed numerical phantom of the thorax, combined with a newly developed algorithm for random selection of electrode positions based on physical limitations resulting from the elasticity of the band and possible position inaccuracies while putting the band on the patient’s chest. The thorax phantom included the heart, lungs, aorta, and spine. Four training and four testing datasets were generated using four different levels of electrode displacement. Reconstruction was conducted using four versions of neural networks trained on the datasets, with random ellipses included and noise added to achieve an SNR of 30 dB. The quality was assessed using pixel-to-pixel metrics such as the root-mean-square error, structural similarity index, 2D correlation coefficient, and peak signal-to-noise ratio. The results showed a strong negative influence of electrode displacement on reconstruction quality when no samples with displaced electrodes were present in the training dataset. Training the network on the dataset containing samples with electrode displacement allowed us to significantly improve the quality of the reconstructed images. Introducing samples with misplaced electrodes increased neural network robustness to electrode displacement while testing. Full article
(This article belongs to the Special Issue State of the Art in Wearable Sensors for Health Monitoring)
►▼ Show Figures

Figure 1

15 pages, 1788 KB  
Article
The Validity and Reliability of the Force Plates and the Linear Position Transducer in Measuring Countermovement Depth and Velocity During Countermovement Jump
by Zheng’ao Li, Wenyue Ma, Ling Zhang, Wenfei Zhu, Qian Xie and Yuliang Sun
Sensors 2025, 25(21), 6542; https://doi.org/10.3390/s25216542 - 23 Oct 2025
Abstract
Countermovement jump (CMJ) is a key test for evaluating lower-limb neuromuscular function, and accurate measurement of countermovement depth (CMD) and countermovement velocity (CMV) is critical for determining optimal performance. However, the measurement validity and reliability of CMD and CMV—particularly when obtained from force [...] Read more.
Countermovement jump (CMJ) is a key test for evaluating lower-limb neuromuscular function, and accurate measurement of countermovement depth (CMD) and countermovement velocity (CMV) is critical for determining optimal performance. However, the measurement validity and reliability of CMD and CMV—particularly when obtained from force plates (FP) and linear position transducers (LPT)—have remained uncertain. This study determined the validity and reliability of FP and LPT for measuring CMD and CMV. Twenty-eight male recreational athletes performed the CMJ test, and the variables were synchronously acquired by Motion Capture (MC), FP, and LPT. The test was divided into two sessions, with participants completing three maximal effort CMJs per session, and the second session occurred more than 48 h after the first. The reliability was evaluated using the intraclass correlation coefficient (ICC), and the validity was evaluated with linear Pearson’s correlation coefficient (r), one-way ANOVA with repeated measures, and Bland–Altman plots. The reliability results for FP and LPT indicated good to excellent (ICC = 0.809–0.900). Compared with MC, the FP showed a high to very high correlation (r = 0.894–0.937), and the LPT showed a high correlation (r = 0.721–0.726). When precise quantification of CMD/CMV is required, FP should be preferred. If only an LPT is available, it is best used for within-athlete longitudinal monitoring with a consistent setup, and cross-device comparisons should be avoided. Full article
(This article belongs to the Special Issue Measurement Sensors and Applications)
►▼ Show Figures

Figure 1

17 pages, 1825 KB  
Article
STCCA: Spatial–Temporal Coupled Cross-Attention Through Hierarchical Network for EEG-Based Speech Recognition
by Liang Dong, Hengyi Shao, Lin Zhang and Lei Li
Sensors 2025, 25(21), 6541; https://doi.org/10.3390/s25216541 - 23 Oct 2025
Abstract
Speech recognition based on Electroencephalogram (EEG) has attracted considerable attention due to its potential in communication and rehabilitation. Existing methods typically process spatial and temporal features with sequential, parallel, or constrained feature fusion strategies. However, the intricate cross-relationships between spatial and temporal features [...] Read more.
Speech recognition based on Electroencephalogram (EEG) has attracted considerable attention due to its potential in communication and rehabilitation. Existing methods typically process spatial and temporal features with sequential, parallel, or constrained feature fusion strategies. However, the intricate cross-relationships between spatial and temporal features remain underexplored. To address these limitations, we propose a spatial–temporal coupled cross-attention mechanism through a hierarchical network, named STCCA. The proposed STCCA consists of three key components: local feature extraction module (LFEM), coupled cross-attention (CCA) fusion module, and global feature extraction module (GFEM). The LFEM employs CNNs to extract local temporal and spatial features, while the CCA fusion module leverages a dual-directional attention mechanism to establish deep interactions between temporal and spatial features. The GFEM uses multi-head self-attention layers to model long-range dependencies and extract global features comprehensively. STCCA is validated on three EEG-based speech datasets, achieving accuracies of 45.45%, 25.91%, and 29.07%, corresponding to improvements of 1.95%, 3.98%, and 1.98% over the comparison models. Full article
(This article belongs to the Special Issue EEG Signal Processing Techniques and Applications—3rd Edition)
►▼ Show Figures

Figure 1

31 pages, 9352 KB  
Systematic Review
Fall Detection in Elderly People: A Systematic Review of Ambient Assisted Living and Smart Home-Related Technology Performance
by Philippe Gorce and Julien Jacquier-Bret
Sensors 2025, 25(21), 6540; https://doi.org/10.3390/s25216540 - 23 Oct 2025
Abstract
Fall detection systems in ambient assisted living (AAL) and smart homes are essential for the comfort, safety, and autonomy of elderly people. The aim of this study was to investigate the performance of these systems considering categories of sensors and methods used. A [...] Read more.
Fall detection systems in ambient assisted living (AAL) and smart homes are essential for the comfort, safety, and autonomy of elderly people. The aim of this study was to investigate the performance of these systems considering categories of sensors and methods used. A systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Seven open databases were screened without a date limit: PubMed/MedLine, Google Scholar, ScienceDirect, Science.gov, Academia, IEEE Xplore, and Mendeley. The article selection and data extraction were performed by two authors independently. Among the 473 unique records, 80 studies were selected. Five fall detection performance parameters (accuracy, precision, sensitivity, specificity, F1-score) and two computation speed parameters (training and testing time) were extracted and classified according to three sensor categories (wearable, non-wearable, and hybrid solutions), and four methods (deep learning, machine learning, threshold, and all others). The ANOVA results showed that wearable sensors performed the worst in fall detection. Deep learning methods produced the best results for the five parameters. Identifying the advantages of different solutions is a major challenge for researchers, practitioners, and policymakers in the design and implementation of more effective fall detection systems. Full article
(This article belongs to the Special Issue Intelligent Sensors and Robots for Ambient Assisted Living)
►▼ Show Figures

Figure 1

52 pages, 5951 KB  
Review
Advanced Metal–Organic Framework-Based Sensor Systems for Gas and Environmental Monitoring: From Material Design to Embedded Applications
by Alemayehu Kidanemariam and Sungbo Cho
Sensors 2025, 25(21), 6539; https://doi.org/10.3390/s25216539 - 23 Oct 2025
Abstract
Environmental pollution is a global issue presenting risks to ecosystems and human health through release of toxic gases, existence of volatile organic compounds (VOCs) in the environment, and heavy metal contamination of waters and soils. To effectively address this issue, reliable and real-time [...] Read more.
Environmental pollution is a global issue presenting risks to ecosystems and human health through release of toxic gases, existence of volatile organic compounds (VOCs) in the environment, and heavy metal contamination of waters and soils. To effectively address this issue, reliable and real-time monitoring technology is imperative. Metal–organic frameworks (MOFs) are a disruptive set of materials with high surface area, tunable porosity, and abundant chemistry to design extremely sensitive and selective pollutant detection. This review article gives an account of recent advances towards sensor technology for MOFs with application specificity towards gas and environment monitoring. We critically examine optical, electrochemical, and resistive platforms and their interfacing with embedded electronics and edge artificial intelligence (edge-AI) to realize smart, compact, and energy-efficient monitoring tools. We also detail critical challenges such as scalability, reproducibility, long-term stability, and secure data management and underscore transforming MOF-based sensors from lab prototype to functional instruments to ensure safe coverage of human health and to bring about sustainable environmental management. Full article
(This article belongs to the Special Issue Advanced Sensors for Gas Monitoring: 2nd Edition)
►▼ Show Figures

Figure 1

15 pages, 3231 KB  
Article
Optimizing Client Participation in Communication-Constrained Federated LLM Adaptation with LoRA
by Faranaksadat Solat and Joohyung Lee
Sensors 2025, 25(21), 6538; https://doi.org/10.3390/s25216538 - 23 Oct 2025
Abstract
Federated learning (FL) enables privacy-preserving adaptation of large language models (LLMs) across distributed clients. However, deploying FL in edge environments remains challenging because of the high communication overhead of full-model updates. Recent advances in parameter-efficient fine-tuning (PEFT), particularly low-rank adaptation (LoRA), have substantially [...] Read more.
Federated learning (FL) enables privacy-preserving adaptation of large language models (LLMs) across distributed clients. However, deploying FL in edge environments remains challenging because of the high communication overhead of full-model updates. Recent advances in parameter-efficient fine-tuning (PEFT), particularly low-rank adaptation (LoRA), have substantially reduced update sizes by injecting lightweight trainable matrices into pretrained transformers, thereby making FL with LLMs more feasible. In this paper, we propose LoRaC-GA, a communication-aware optimization framework that dynamically determines the optimal number of clients to participate in each round under a fixed bandwidth constraint. We formulated a max-min objective to jointly maximize the model accuracy and communication efficiency and solved the resulting non-convex problem using a genetic algorithm (GA). To further reduce the overhead, we integrated a structured peer-to-peer collaboration protocol with log2K complexity, enabling scalable communication without full connectivity. The simulation results demonstrate that LoRaC-GA adaptively selects the optimal client count, achieving competitive accuracy while significantly reducing the communication cost. The proposed framework is well-suited for bandwidth-constrained edge deployments involving large-scale LLMs. Full article
(This article belongs to the Special Issue Artificial Intelligence (AI) and Cyber–Physical–Social Sensing (CPSS))
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-65
Previous Issue
Back to TopTop