Next Issue
Volume 25, July-1
Previous Issue
Volume 25, June-1
 
 
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 12
sensors-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Sensors, Volume 25, Issue 12 (June-2 2025) – 264 articles

Cover Story (view full-size image): Gas-sensing technologies play a vital role across diverse fields, ranging from environmental monitoring and industrial process control to medical diagnostics and security applications. The integration of photonic–piezoelectric components is a key factor in offering rugged and portable sensing devices. This research demonstrates a compact optical gas-sensing system integrating silicon nitride waveguide with a custom-designed quartz tuning fork for trace gas detection. The system employs both quartz-enhanced photoacoustic spectroscopy and light-induced thermoelastic spectroscopy techniques, achieving comparable signal-to-noise ratios when detecting 1.6% water vapor concentration. The compact nature of this architecture also opens the way to multiplexed sensing via integrating multiple waveguides and laser sources on a single chip. 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:
22 pages, 8644 KiB  
Article
Privacy-Preserving Approach for Early Detection of Long-Lie Incidents: A Pilot Study with Healthy Subjects
by Riska Analia, Anne Forster, Sheng-Quan Xie and Zhiqiang Zhang
Sensors 2025, 25(12), 3836; https://doi.org/10.3390/s25123836 - 19 Jun 2025
Viewed by 334
Abstract
(1) Background: Detecting long-lie incidents—where individuals remain immobile after a fall—is essential for timely intervention and preventing severe health consequences. However, most existing systems focus only on fall detection, neglect post-fall monitoring, and raise privacy concerns, especially in real-time, non-invasive applications; (2) Methods: [...] Read more.
(1) Background: Detecting long-lie incidents—where individuals remain immobile after a fall—is essential for timely intervention and preventing severe health consequences. However, most existing systems focus only on fall detection, neglect post-fall monitoring, and raise privacy concerns, especially in real-time, non-invasive applications; (2) Methods: This study proposes a lightweight, privacy-preserving, long-lie detection system utilizing thermal imaging and a soft-voting ensemble classifier. A low-resolution thermal camera captured simulated falls and activities of daily living (ADL) performed by ten healthy participants. Human pose keypoints were extracted using MediaPipe, followed by the computation of five handcrafted postural features. The top three classifiers—automatically selected based on cross-validation performance—formed the soft-voting ensemble. Long-lie conditions were identified through post-fall immobility monitoring over a defined period, using rule-based logic on posture stability and duration; (3) Results: The ensemble model achieved high classification performance with accuracy, precision, recall, and an F1 score of 0.98. Real-time deployment on a Raspberry Pi 5 demonstrated the system is capable of accurately detecting long-lie incidents based on continuous monitoring over 15 min, with minimal posture variation; (4) Conclusion: The proposed system introduces a novel approach to long-lie detection by integrating privacy-aware sensing, interpretable posture-based features, and efficient edge computing. It demonstrates strong potential for deployment in homecare settings. Future work includes validation with older adults and integration of vital sign monitoring for comprehensive assessment. Full article
(This article belongs to the Special Issue Sensing and Signal Processing Technologies for Outpatient Monitoring and Rehabilitation)
►▼ Show Figures

Figure 1

14 pages, 3205 KiB  
Article
A 209 ps Shutter-Time CMOS Image Sensor for Ultra-Fast Diagnosis
by Houzhi Cai, Zhaoyang Xie, Youlin Ma and Lijuan Xiang
Sensors 2025, 25(12), 3835; https://doi.org/10.3390/s25123835 - 19 Jun 2025
Viewed by 214
Abstract
A conventional microchannel plate framing camera is typically utilized for inertial confinement fusion diagnosis. However, as a vacuum electronic device, it has inherent limitations, such as a complex structure and the inability to achieve single-line-of-sight imaging. To address these challenges, a CMOS image [...] Read more.
A conventional microchannel plate framing camera is typically utilized for inertial confinement fusion diagnosis. However, as a vacuum electronic device, it has inherent limitations, such as a complex structure and the inability to achieve single-line-of-sight imaging. To address these challenges, a CMOS image sensor that can be seamlessly integrated with an electronic pulse broadening system can provide a viable alternative to the microchannel plate detector. This paper introduces the design of an 8 × 8 pixel-array ultrashort shutter-time single-framing CMOS image sensor, which leverages silicon epitaxial processing and a 0.18 μm standard CMOS process. The focus of this study is on the photodiode and the readout pixel-array circuit. The photodiode, designed using the silicon epitaxial process, achieves a quantum efficiency exceeding 30% in the visible light band at a bias voltage of 1.8 V, with a temporal resolution greater than 200 ps for visible light. The readout pixel-array circuit, which is based on the 0.18 μm standard CMOS process, incorporates 5T structure pixel units, voltage-controlled delayers, clock trees, and row-column decoding and scanning circuits. Simulations of the pixel circuit demonstrate an optimal temporal resolution of 60 ps. Under the shutter condition with the best temporal resolution, the maximum output swing of the pixel circuit is 448 mV, and the output noise is 77.47 μV, resulting in a dynamic range of 75.2 dB for the pixel circuit; the small-signal responsivity is 1.93 × 10−7 V/e−, and the full-well capacity is 2.3 Me−. The maximum power consumption of the 8 × 8 pixel-array and its control circuits is 0.35 mW. Considering both the photodiode and the pixel circuit, the proposed CMOS image sensor achieves a temporal resolution better than 209 ps. Full article
(This article belongs to the Special Issue Ultrafast Optoelectronic Sensing and Imaging)
►▼ Show Figures

Figure 1

13 pages, 429 KiB  
Article
Comparative Analysis of In-Match Physical Requirements Across National and International Competitive Contexts in Cerebral Palsy Football
by Juan Francisco Maggiolo, Juan José García-Hernández, Manuel Moya-Ramón and Iván Peña-González
Sensors 2025, 25(12), 3834; https://doi.org/10.3390/s25123834 - 19 Jun 2025
Viewed by 179
Abstract
This study aimed to compare in-match physical and technical requirements of cerebral palsy (CP) football players across different national and international competitive contexts. A total of 79 male outfield players participated in 62 official matches across 3 competitive phases of the Spanish National [...] Read more.
This study aimed to compare in-match physical and technical requirements of cerebral palsy (CP) football players across different national and international competitive contexts. A total of 79 male outfield players participated in 62 official matches across 3 competitive phases of the Spanish National CP Football League (Regular Phase, Consolation Phase, and Playoffs) and the IFCPF World Cup. Inertial measurement units (IMUs) were used to record locomotor and technical variables during each match. A subset of 10 players was tracked across all phases. Physical demands were normalized per minute of play and analyzed using one-way and repeated-measures ANOVAs. Results revealed that physical requirements during the World Cup were up to three times higher than during national-level matches, with significantly greater maximum velocities, high-intensity distances, and frequencies of accelerations and decelerations (p < 0.001, ηp2 > 0.40). Playoffs also imposed significantly greater physical requirements compared to Regular and Consolation Phases. International matches showed a markedly higher number of ball contacts, indicating increased technical involvement. These patterns were consistent in both the full sample and the longitudinal subsample, suggesting that competitive level—rather than player characteristics alone— strongly modulates physical output during the competition. These findings underscore the need for context-specific training and load management strategies to prepare athletes for the elevated demands of high-level CP football competition. Full article
(This article belongs to the Section Wearables)
►▼ Show Figures

Figure 1

13 pages, 1696 KiB  
Article
Commercial Hoverboard Reverse Engineering and Repurposing for a Stabilized Platform: A Recyclable Solution for Modular Robotic Bases
by Antoine Leblanc, Lùka Tricot, Duncan Briquet, Mohamed Aziz Slama and Christophe Delebarre
Sensors 2025, 25(12), 3833; https://doi.org/10.3390/s25123833 - 19 Jun 2025
Viewed by 257
Abstract
Sustainability and resource optimization have spurred interest in giving a second life to used equipment, often discarded after limited use. Within this framework, we conducted a multidisciplinary, final-year engineering project to explore the reverse engineering and repurposing of commercial hoverboards for an auto-stabilizing, [...] Read more.
Sustainability and resource optimization have spurred interest in giving a second life to used equipment, often discarded after limited use. Within this framework, we conducted a multidisciplinary, final-year engineering project to explore the reverse engineering and repurposing of commercial hoverboards for an auto-stabilizing, modular robotic platform, with emphasis on medical applications such as transporting medication. The innovation lies in recycling hoverboards to develop a teleoperated, stabilized base that can accommodate additional modules—for instance, a multifunctional arm or a transport shelf—akin to existing commercial robots. Our methodology involves disassembling and reprogramming the hoverboard’s motor controllers and sensors to maintain horizontal stability. Control is realized through the sensor fusion of accelerometer and gyroscope data, processed by a Kalman filter and implemented in a Proportional-Integral-Derivative (PID) loop. A user-friendly Human-Machine Interface (HMI), hosted on an ESP32 microcontroller, enables remote operation and monitoring. Experimental results show that the platform autonomously balances, carries payloads, and achieves high energy efficiency, highlighting its potential as a sustainable and versatile solution in modular robotic applications. Full article
(This article belongs to the Section Sensors and Robotics)
►▼ Show Figures

Figure 1

18 pages, 2873 KiB  
Article
Enhanced AlexNet with Gabor and Local Binary Pattern Features for Improved Facial Emotion Recognition
by Furkat Safarov, Alpamis Kutlimuratov, Ugiloy Khojamuratova, Akmalbek Abdusalomov and Young-Im Cho
Sensors 2025, 25(12), 3832; https://doi.org/10.3390/s25123832 - 19 Jun 2025
Viewed by 293
Abstract
Facial emotion recognition (FER) is vital for improving human–machine interactions, serving as the foundation for AI systems that integrate cognitive and emotional intelligence. This helps bridge the gap between mechanical processes and human emotions, enhancing machine engagement with humans. Considering the constraints of [...] Read more.
Facial emotion recognition (FER) is vital for improving human–machine interactions, serving as the foundation for AI systems that integrate cognitive and emotional intelligence. This helps bridge the gap between mechanical processes and human emotions, enhancing machine engagement with humans. Considering the constraints of low hardware specifications often encountered in real-world applications, this study leverages recent advances in deep learning to propose an enhanced model for FER. The model effectively utilizes texture information from faces through Gabor and Local Binary Pattern (LBP) feature extraction techniques. By integrating these features into a specially modified AlexNet architecture, our approach not only classifies facial emotions more accurately but also demonstrates significant improvements in performance and adaptability under various operational conditions. To validate the effectiveness of our proposed model, we conducted evaluations using the FER2013 and RAF-DB benchmark datasets, where it achieved impressive accuracies of 98.10% and 93.34% for the two datasets, with standard deviations of 1.63% and 3.62%, respectively. On the FER-2013 dataset, the model attained a precision of 98.2%, a recall of 97.9%, and an F1-score of 98.0%. Meanwhile, for the other dataset, it achieved a precision of 93.54%, a recall of 93.12%, and an F1-score of 93.34%. These results underscore the model’s robustness and its capability to deliver high-precision emotion recognition, making it an ideal solution for deployment in environments where hardware limitations are a critical concern. Full article
(This article belongs to the Special Issue Computer Vision and Machine Learning for Intelligent Sensing Systems—2nd Edition)
►▼ Show Figures

Figure 1

16 pages, 545 KiB  
Article
Microcontroller Implementation of LSTM Neural Networks for Dynamic Hand Gesture Recognition
by Kevin Di Leo, Giorgio Biagetti, Laura Falaschetti and Paolo Crippa
Sensors 2025, 25(12), 3831; https://doi.org/10.3390/s25123831 - 19 Jun 2025
Viewed by 210
Abstract
Accelerometers are nowadays included in almost any portable or mobile device, including smartphones, smartwatches, wrist-bands, and even smart rings. The data collected from them is therefore an ideal candidate to tackle human motion recognition, as it can easily and unobtrusively be acquired. In [...] Read more.
Accelerometers are nowadays included in almost any portable or mobile device, including smartphones, smartwatches, wrist-bands, and even smart rings. The data collected from them is therefore an ideal candidate to tackle human motion recognition, as it can easily and unobtrusively be acquired. In this work we analyze the performance of a hand-gesture classification system implemented using LSTM neural networks on a resource-constrained microcontroller platform, which required trade-offs between network accuracy and resource utilization. Using a publicly available dataset, which includes data for 20 different hand gestures recorded from 10 subjects using a wrist-worn device with a 3-axial accelerometer, we achieved nearly 90.25% accuracy while running the model on an STM32L4-series microcontroller, with an inference time of 418 ms for 4 s sequences, corresponding to an average CPU usage of about 10% for the recognition task. Full article
(This article belongs to the Special Issue Smart Sensing Technology for Human Activity Recognition)
►▼ Show Figures

Figure 1

16 pages, 3028 KiB  
Article
Multi-Modal Joint Pulsed Eddy Current Sensor Signal Denoising Method Integrating Inductive Disturbance Mechanism
by Yun Zuo, Gebiao Hu, Fan Gan, Zhiwu Zeng, Zhichi Lin, Xinxun Wang, Ruiqing Xu, Liang Wen, Shubing Hu, Haihong Le, Runze Wu and Jingang Wang
Sensors 2025, 25(12), 3830; https://doi.org/10.3390/s25123830 - 19 Jun 2025
Viewed by 240
Abstract
Pulsed eddy current (PEC) testing technology has been widely used in the field of non-destructive testing of metal grounding structures due to its wide-band excitation and response characteristics. However, multi-source noise in industrial environments can significantly degrade the performance of PEC sensors, thereby [...] Read more.
Pulsed eddy current (PEC) testing technology has been widely used in the field of non-destructive testing of metal grounding structures due to its wide-band excitation and response characteristics. However, multi-source noise in industrial environments can significantly degrade the performance of PEC sensors, thereby limiting their detection accuracy. This study proposes a multi-modal joint pulsed eddy current signal sensor denoising method that integrates the inductive disturbance mechanism. This method constructs the Improved Whale Optimization -Variational Mode Decomposition-Singular Value Decomposition-Wavelet Threshold Denoising (IWOA-VMD-SVD-WTD) fourth-order processing architecture: IWOA adaptively optimizes the VMD essential variables (K, α) and employs the optimized VMD to decompose the perception coefficient (IMF) of the PEC signal. It utilizes the correlation coefficient criterion to filter and identify the primary noise components within the signal, and the SVD-WTD joint denoising model is established to reconstruct each component to remove the noise signal received by the PEC sensor. To ascertain the efficacy of this approach, we compared the IWOA-VMD-SVD-WTD method with other denoising methods under three different noise levels through experiments. The test results show that compared with other VMD-based denoising techniques, the average signal-to-noise ratio (SNR) of the PEC signal received by the receiving coil for 200 noise signals in different noise environments is 24.31 dB, 29.72 dB and 29.64 dB, respectively. The average SNR of the other two denoising techniques in different noise environments is 15.48 dB, 18.87 dB, 18.46 dB and 19.32 dB, 27.13 dB, 26.78 dB, respectively, which is significantly better than other denoising methods. In addition, in practical applications, this method is better than other technologies in denoising PEC signals and successfully achieves noise reduction and signal feature extraction. This study provides a new technical solution for extracting pure and impurity-free PEC signals in complex electromagnetic environments. Full article
(This article belongs to the Section Industrial Sensors)
►▼ Show Figures

Figure 1

20 pages, 3616 KiB  
Article
Res-RBG Facial Expression Recognition in Image Sequences Based on Dual Neural Networks
by Xiangwei Mou, Yongfu Song, Xiuping Xie, Mingxuan You and Rijun Wang
Sensors 2025, 25(12), 3829; https://doi.org/10.3390/s25123829 - 19 Jun 2025
Viewed by 163
Abstract
Facial expressions involve dynamic changes, and facial expression recognition based on static images struggles to capture the temporal information inherent in these dynamic changes. The resultant degradation in real-world performance critically impedes the integration of facial expression recognition systems into intelligent sensing applications. [...] Read more.
Facial expressions involve dynamic changes, and facial expression recognition based on static images struggles to capture the temporal information inherent in these dynamic changes. The resultant degradation in real-world performance critically impedes the integration of facial expression recognition systems into intelligent sensing applications. Therefore, this paper proposes a facial expression recognition method for image sequences based on the fusion of dual neural networks (ResNet and residual bidirectional GRU—Res-RBG). The model proposed in this paper achieves recognition accuracies of 98.10% and 88.64% on the CK+ and Oulu-CASIA datasets, respectively. Moreover, the model has a parameter size of only 64.20 M. Compared to existing methods for image sequence-based facial expression recognition, the approach presented in this paper demonstrates certain advantages, indicating strong potential for future edge sensor deployment. Full article
(This article belongs to the Special Issue Sensing Technologies Applied in Human Emotion and Facial Expression Recognition)
►▼ Show Figures

Figure 1

25 pages, 28388 KiB  
Article
Software Trusted Platform Module (SWTPM) Resource Sharing Scheme for Embedded Systems
by Da-Chuan Chen, Guan-Ruei Chen and Yu-Ping Liao
Sensors 2025, 25(12), 3828; https://doi.org/10.3390/s25123828 - 19 Jun 2025
Viewed by 227
Abstract
Embedded system networks are widely deployed across various domains and often perform mission-critical tasks, making it essential for all nodes within the system to be trustworthy. Traditionally, each node is equipped with a discrete Trusted Platform Module (dTPM) to ensure network-wide trustworthiness. However, [...] Read more.
Embedded system networks are widely deployed across various domains and often perform mission-critical tasks, making it essential for all nodes within the system to be trustworthy. Traditionally, each node is equipped with a discrete Trusted Platform Module (dTPM) to ensure network-wide trustworthiness. However, this study proposes a cost-effective system architecture that deploys software-based TPMs (SWTPMs) on the majority of nodes, while reserving dTPMs for a few central nodes to maintain overall system integrity. The proposed architecture employs IBMACS for system integrity reporting. In addition, a database-based anomaly detection (AD) agent is developed to identify and isolate untrusted nodes. A traffic anomaly detection agent is also introduced to monitor communication between servers and clients, ensuring that traffic patterns remain normal. Finally, a custom measurement kernel is implemented, along with an activation agent, to enforce a measured boot process for custom applications during startup. This architecture is designed to safeguard mission-critical embedded systems from malicious threats while reducing deployment costs. Full article
(This article belongs to the Special Issue Privacy and Security for IoT-Based Smart Homes)
►▼ Show Figures

Figure 1

19 pages, 2488 KiB  
Article
An Improved Segformer for Semantic Segmentation of UAV-Based Mine Restoration Scenes
by Feng Wang, Lizhuo Zhang, Tao Jiang, Zhuqi Li, Wangyu Wu and Yingchun Kuang
Sensors 2025, 25(12), 3827; https://doi.org/10.3390/s25123827 - 19 Jun 2025
Viewed by 222
Abstract
Mine ecological restoration is a critical process for promoting the sustainable development of resource-dependent regions, yet existing monitoring methods remain limited in accuracy and adaptability. To address challenges such as small-object recognition, insufficient multi-scale feature fusion, and blurred boundaries in UAV-based remote sensing [...] Read more.
Mine ecological restoration is a critical process for promoting the sustainable development of resource-dependent regions, yet existing monitoring methods remain limited in accuracy and adaptability. To address challenges such as small-object recognition, insufficient multi-scale feature fusion, and blurred boundaries in UAV-based remote sensing imagery, this paper proposes an enhanced semantic segmentation model based on Segformer. Specifically, a multi-scale feature-enhanced feature pyramid network (MSFE-FPN) is introduced between the encoder and decoder to strengthen cross-level feature interaction. Additionally, a selective feature aggregation pyramid pooling module (SFA-PPM) is integrated into the deepest feature layer to improve global semantic perception, while an efficient local attention (ELA) module is embedded into lateral connections to enhance sensitivity to edge structures and small-scale targets. A high-resolution UAV image dataset, named the HUNAN Mine UAV Dataset (HNMUD), is constructed to evaluate model performance, and further validation is conducted on the public Aeroscapes dataset. Experimental results demonstrated that the proposed method exhibited strong performance in terms of segmentation accuracy and generalization ability, effectively supporting the image analysis needs of mine restoration scenes. Full article
(This article belongs to the Section Remote Sensors)
►▼ Show Figures

Figure 1

21 pages, 4676 KiB  
Article
RFID-Based Real-Time Salt Concentration Monitoring with Adaptive EKF
by Renhai Feng and Xinyi Lin
Sensors 2025, 25(12), 3826; https://doi.org/10.3390/s25123826 - 19 Jun 2025
Viewed by 250
Abstract
Salt concentration monitoring is crucial for industrial process control and wastewater management, yet existing methods often lack real-time capability or require invasive sampling. This paper presents a novel RFID wireless sensing system for noninvasive solution concentration monitoring, combining physical modeling with advanced estimation [...] Read more.
Salt concentration monitoring is crucial for industrial process control and wastewater management, yet existing methods often lack real-time capability or require invasive sampling. This paper presents a novel RFID wireless sensing system for noninvasive solution concentration monitoring, combining physical modeling with advanced estimation algorithms. By combining the Cole–Cole model and the slit cylindrical capacitor (SCC) model, the system establishes physics-based state-space models to characterize concentration-dependent RFID signal variations. The concentration dynamics are modeled as a hidden Markov process and tracked using an adaptive extended Kalman filter (AEKF). The AEKF algorithm avoids computationally expensive inversion of complex observation equations while automatically adjusting noise covariance matrices via innovation sequence. Experimental results demonstrate a mean relative error (MRE) of 2.8% for CaCl2 solution across 2–10 g/L concentrations. Within the experimentally validated optimal range (2–8 g/L CaCl2), the system maintains MRE below 3% under artificially introduced measurement noise, confirming its strong robustness. Compared with baseline approaches, the proposed AEKF algorithm shows improved performance in both accuracy and computational efficiency. Full article
(This article belongs to the Section Environmental Sensing)
►▼ Show Figures

Figure 1

31 pages, 1107 KiB  
Article
Length–Weight Distribution of Non-Zero Elements in Randomized Bit Sequences
by Christoph Lange, Andreas Ahrens, Yadu Krishnan Krishnakumar and Olaf Grote
Sensors 2025, 25(12), 3825; https://doi.org/10.3390/s25123825 - 19 Jun 2025
Viewed by 220
Abstract
Randomness plays an important role in data communication as well as in cybersecurity. In the simulation of communication systems, randomized bit sequences are often used to model a digital source information stream. Cryptographic outputs should look more random than deterministic in order to [...] Read more.
Randomness plays an important role in data communication as well as in cybersecurity. In the simulation of communication systems, randomized bit sequences are often used to model a digital source information stream. Cryptographic outputs should look more random than deterministic in order to provide an attacker with as little information as possible. Therefore, the investigation of randomness, especially in cybersecurity, has attracted a lot of attention and research activities. Common tests regarding randomness are hypothesis-based and focus on analyzing the distribution and independence of zero and non-zero elements in a given random sequence. In this work, a novel approach grounded in a gap-based burst analysis is presented and analyzed. Such approaches have been successfully implemented, e.g., in data communication systems and data networks. The focus of the current work is on detecting deviations from the ideal gap-density function describing randomized bit sequences. For testing and verification purposes, the well-researched post-quantum cryptographic CRYSTALS suite, including its Kyber and Dilithium schemes, is utilized. The proposed technique allows for quickly verifying the level of randomness in given cryptographic outputs. The results for different sequence-generation techniques are presented, thus validating the approach. The results show that key-encapsulation and key-exchange algorithms, such as CRYSTALS-Kyber, achieve a lower level of randomness compared to digital signature algorithms, such as CRYSTALS-Dilithium. Full article
(This article belongs to the Section Communications)
►▼ Show Figures

Figure 1

13 pages, 2599 KiB  
Article
Fiber-Coupled Multipass NIR Sensor for In Situ, Real-Time Water Vapor Outgassing Monitoring
by Logan Echeveria, Yue Hao, Michael C. Rushford, Gerardo Chavez, Sean Tardif, Allan Chang, Sylvie Aubry, Maxwell Murialdo, J. Chance Carter, Brandon Foley, Pratanu Roy, S. Roger Qiu and Tiziana Bond
Sensors 2025, 25(12), 3824; https://doi.org/10.3390/s25123824 - 19 Jun 2025
Viewed by 244
Abstract
This work presents the recent development of a fiber-coupled multipass near-infrared (NIR) gas sensor used to monitor water vapor desorption of small material coupons. The gas sensor design employs a White cell topology to maximize the optical path length over a compact, hand-size [...] Read more.
This work presents the recent development of a fiber-coupled multipass near-infrared (NIR) gas sensor used to monitor water vapor desorption of small material coupons. The gas sensor design employs a White cell topology to maximize the optical path length over a compact, hand-size footprint. Water vapor concentrations are quantified over a large dynamic range by simultaneously applying wavelength modulation and tunable diode laser absorption spectroscopy techniques. A custom headspace optimized for material desorption experiments is assembled using commercially available vacuum chamber components. We provide in situ measurements of water vapor desorption from two geometries of the industrially important silicone elastomer Sylgard-184 as a case study for sensor viability. To corroborate the results, the gas sensor data are compared to numerical simulations based on a triple-mode diffusion–sorption model, consisting of Henry, Langmuir, and Pooling modes. Full article
(This article belongs to the Special Issue Optical Sensors for Industrial Applications)
►▼ Show Figures

Figure 1

22 pages, 2651 KiB  
Article
Multi-Party Verifiably Collaborative Encryption for Biomedical Signals via Singular Spectrum Analysis-Based Chaotic Filter Bank Networks
by Xiwen Zhang, Jianfeng He and Bingo Wing-Kuen Ling
Sensors 2025, 25(12), 3823; https://doi.org/10.3390/s25123823 - 19 Jun 2025
Viewed by 147
Abstract
This paper proposes a multi-party verifiably collaborative system for encrypting the nonlinear and the non-stationary biomedical signals captured by biomedical sensors via the singular spectrum analysis (SSA)-based chaotic networks. In particular, the raw signals are first decomposed into the multiple components by the [...] Read more.
This paper proposes a multi-party verifiably collaborative system for encrypting the nonlinear and the non-stationary biomedical signals captured by biomedical sensors via the singular spectrum analysis (SSA)-based chaotic networks. In particular, the raw signals are first decomposed into the multiple components by the SSA. Then, these decomposed components are fed into the chaotic filter bank networks for performing the encryption. To perform the multi-party verifiably collaborative encryption, the window length of the SSA and the total number of the layers in the chaotic network are flexibly designed to match the total number of the collaborators. The computer numerical simulation results show that our proposed system achieves a good encryption performance. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning Approaches for Processing Biomedical Sensor Signals)
►▼ Show Figures

Figure 1

22 pages, 969 KiB  
Article
A Spectral Interpretable Bearing Fault Diagnosis Framework Powered by Large Language Models
by Panfeng Bao, Wenjun Yi, Yue Zhu, Yufeng Shen and Haotian Peng
Sensors 2025, 25(12), 3822; https://doi.org/10.3390/s25123822 - 19 Jun 2025
Viewed by 311
Abstract
Most existing fault diagnosis methods, although capable of extracting interpretable features such as attention-weighted fault-related frequencies, remain essentially black-box models that provide only classification results without transparent reasoning or diagnostic justification, limiting users’ ability to understand and trust diagnostic outcomes. In this work, [...] Read more.
Most existing fault diagnosis methods, although capable of extracting interpretable features such as attention-weighted fault-related frequencies, remain essentially black-box models that provide only classification results without transparent reasoning or diagnostic justification, limiting users’ ability to understand and trust diagnostic outcomes. In this work, we present a novel, interpretable fault diagnosis framework that integrates spectral feature extraction with large language models (LLMs). Vibration signals are first transformed into spectral representations using Hilbert- and Fourier-based encoders to highlight key frequencies and amplitudes. A channel attention-augmented convolutional neural network provides an initial fault type prediction. Subsequently, structured information—including operating conditions, spectral features, and CNN outputs—is fed into a fine-tuned enhanced LLM, which delivers both an accurate diagnosis and a transparent reasoning process. Experiments demonstrate that our framework achieves high diagnostic performance while substantially improving interpretability, making advanced fault diagnosis accessible to non-expert users in industrial settings. Full article
(This article belongs to the Topic Advances in Integrative AI, Machine Learning, and Big Data for Transformative Applications)
►▼ Show Figures

Figure 1

28 pages, 280 KiB  
Review
Research Progress and Prospects of Intelligent Measurement and Control Technology for Tillage Depth in Subsoiling Operations
by Yue Deng, Wenyi Zhang, Bing Qi, Yunxia Wang, Youqiang Ding and Haojie Zhang
Sensors 2025, 25(12), 3821; https://doi.org/10.3390/s25123821 - 19 Jun 2025
Viewed by 249
Abstract
Deep tillage is a conservation tillage method that aims to break the plow pan layer. It provides significant benefits, including enhanced root development, improved soil quality, and substantial increases in crop yields. The depth of tillage is a crucial factor in assessing the [...] Read more.
Deep tillage is a conservation tillage method that aims to break the plow pan layer. It provides significant benefits, including enhanced root development, improved soil quality, and substantial increases in crop yields. The depth of tillage is a crucial factor in assessing the effectiveness of deep tillage operations. Accurate regulation of tillage depth in deep tillage equipment is vital for ensuring the high-quality and efficient execution of these practices. The distribution of mechanical resistance within the soil can effectively indicate the location of the plow pan layer and serves as the main reference for setting the tillage depth for machinery. This paper examined the current state of research on tillage depth control technology for deep tillage operations. It focused on three main technical areas: soil mechanical resistance detection, tillage depth measurement, and tillage depth regulation. The report discussed the working principles of various technologies and compared the existing methods. Additionally, the paper analyzed the challenges faced in the development of tillage depth control technology in China and offers recommendations for future advancements. It highlighted that leveraging information and digital technologies to determine the distribution of the soil plow pan layer, along with the integration of efficient and intelligent control technologies for precise tillage depth regulation, represented a key direction for the future development of deep tillage operations. Full article
(This article belongs to the Special Issue Smart Sensors in Precision Agriculture)
25 pages, 5546 KiB  
Article
A Portable Insole System for Actively Controlled Offloading of Plantar Pressure for Diabetic Foot Care
by Pedro Castro-Martins, Arcelina Marques, Luís Pinto-Coelho, Pedro Fonseca and Mário Vaz
Sensors 2025, 25(12), 3820; https://doi.org/10.3390/s25123820 - 19 Jun 2025
Viewed by 287
Abstract
Plantar pressure monitoring is decisive in injury prevention, especially in at-risk populations such as people with diabetic foot. In this context, innovative solutions such as pneumatic insoles can be essential in plantar pressure management. This study describes the development of a variable pressure [...] Read more.
Plantar pressure monitoring is decisive in injury prevention, especially in at-risk populations such as people with diabetic foot. In this context, innovative solutions such as pneumatic insoles can be essential in plantar pressure management. This study describes the development of a variable pressure system that promotes the monitoring, stabilization, and offloading of plantar pressure through a pneumatic insole. This research was also intended to evaluate its ability to redistribute plantar pressure, reduce peak pressure in both static and dynamic conditions, and validate its pressure measurements by comparing the results with those obtained from a pedar® insole. Tests were carried out under both static and dynamic conditions, before and after the pressure stabilization process by air cells and the subsequent pressure offloading. During the validation process, methods were used to evaluate the agreement between measurements obtained by the two systems. The results of the static test showed that pressure stabilization reduced pressure on the heel by 32.43%, distributing it to the metatarsals and toes. After heel pressure offloading, the reduction reached 42.72%. In the dynamic test, despite natural dispersion of the measurements, a trend to reduce the peak pressure in the heel, metatarsals, and toes was observed. Agreement analysis recorded 96.32% in the static test and 94.02% in the dynamic test. The pneumatic insole proved effective in redistributing and reducing plantar pressure, with more evident effects in the static test. Its agreement with the pedar® system reinforces its reliability as a tool for measuring and managing plantar pressure, representing a promising solution for preventing plantar lesions. Full article
(This article belongs to the Special Issue Sensors and Imaging Technologies for Foot Health Monitoring and Rehabilitation)
►▼ Show Figures

Figure 1

19 pages, 2216 KiB  
Article
Research on Time Constant Test of Thermocouples Based on QNN-PID Controller
by Chenyang Xu, Xiaojian Hao, Pan Pei, Tong Wei and Shenxiang Feng
Sensors 2025, 25(12), 3819; https://doi.org/10.3390/s25123819 - 19 Jun 2025
Viewed by 224
Abstract
The aim of this study is to solve the problem of it being difficult to obtain quantitative step signals when testing the time constant of thermocouples using the laser excitation method, thereby restricting the accuracy and repeatability of the test of the time [...] Read more.
The aim of this study is to solve the problem of it being difficult to obtain quantitative step signals when testing the time constant of thermocouples using the laser excitation method, thereby restricting the accuracy and repeatability of the test of the time constant of thermocouples. This paper designs a thermocouple time constant testing system in which laser power can be adjusted in real time. The thermocouple to be tested and a colorimetric thermometer with a faster response speed are placed on a pair of conjugate focal points of an elliptic mirror. By taking advantage of the aberration-free imaging characteristic of the conjugate focus, the temperature measured by the colorimetric thermometer is taken as the true value on the surface of the thermocouple so as to adjust the output power of the laser in real time, make the output curve of the thermocouple reach a steady state, and calculate the time constant of the thermocouple. This paper simulates and analyzes the effects of adjusting PID parameters using quantum neural networks. By comparing this with the method of optimizing PID parameters with BP neural networks, the superiority of the designed QNN-PID controller is proven. The designed controller was applied to the test system, and the dynamic response curves of the thermocouple reaching equilibrium at the expected temperatures of 800 °C, 900 °C, 1000 °C, 1050 °C, and 1100 °C were obtained. Through calculation, it was obtained that the time constants of the tested thermocouples were all within 150 ms, proving that this system can be used for the time constant test of rapid thermocouples. This also provides a basis for the selection of thermocouples in other subsequent temperature tests. Meanwhile, repeated experiments were conducted on the thermocouple test system at 1000 °C, once again verifying the feasibility of the test system and the repeatability of the experiment. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

17 pages, 2093 KiB  
Article
The Reliability and Validity of an Instrumented Device for Tracking the Shoulder Range of Motion
by Rachel E. Roos, Jennifer Lambiase, Michelle Riffitts, Leslie Scholle, Simran Kulkarni, Connor L. Luck, Dharma Parmanto, Vayu Putraadinatha, Made D. Yoga, Stephany N. Lang, Erica Tatko, Jim Grant, Jennifer I. Oakley, Ashley Disantis, Andi Saptono, Bambang Parmanto, Adam Popchak, Michael P. McClincy and Kevin M. Bell
Sensors 2025, 25(12), 3818; https://doi.org/10.3390/s25123818 - 18 Jun 2025
Viewed by 269
Abstract
Rotator cuff tears are common in individuals over 40, and physical therapy is often prescribed post-surgery. However, access can be limited by cost, convenience, and insurance coverage. CuffLink is a telehealth rehabilitation system that integrates the Strengthening and Stabilization System mechanical exerciser with [...] Read more.
Rotator cuff tears are common in individuals over 40, and physical therapy is often prescribed post-surgery. However, access can be limited by cost, convenience, and insurance coverage. CuffLink is a telehealth rehabilitation system that integrates the Strengthening and Stabilization System mechanical exerciser with the interACTION mobile health platform. The system includes a triple-axis accelerometer (LSM6DSOX + LIS3MDL FeatherWing), a rotary encoder, a VL530X time-of-flight sensor, and two wearable BioMech Health IMUs to capture upper-limb motion. CuffLink is designed to facilitate controlled, home-based exercise while enabling clinicians to remotely monitor joint function. Concurrent validity and test–retest reliability were used to assess device accuracy and repeatability. The results showed moderate to good validity for shoulder rotation (ICC = 0.81), device rotation (ICC = 0.94), and linear tracking (from zero: ICC = 0.75 and RMSE = 2.41; from start: ICC = 0.88 and RMSE = 2.02) and good reliability (e.g., RMSEs as low as 1.66 cm), with greater consistency in linear tracking compared to angular measures. Shoulder rotation and abduction exhibited higher variability in both validity and reliability measures. Future improvements will focus on manufacturability, signal stability, and force sensing. CuffLink supports accessible, data-driven rehabilitation and holds promise for advancing digital health in orthopedic recovery. Full article
(This article belongs to the Special Issue IMU and Innovative Sensors for Healthcare)
►▼ Show Figures

Figure 1

26 pages, 4992 KiB  
Article
NDVI and Beyond: Vegetation Indices as Features for Crop Recognition and Segmentation in Hyperspectral Data
by Andreea Nițu, Corneliu Florea, Mihai Ivanovici and Andrei Racoviteanu
Sensors 2025, 25(12), 3817; https://doi.org/10.3390/s25123817 - 18 Jun 2025
Viewed by 187
Abstract
Vegetation indices have long been central to vegetation monitoring through remote sensing. The most popular one is the Normalized Difference Vegetation Index (NDVI), yet many vegetation indices (VIs) exist. In this paper, we investigate their distinctiveness and discriminative power in the context of [...] Read more.
Vegetation indices have long been central to vegetation monitoring through remote sensing. The most popular one is the Normalized Difference Vegetation Index (NDVI), yet many vegetation indices (VIs) exist. In this paper, we investigate their distinctiveness and discriminative power in the context of applications for agriculture based on hyperspectral data. More precisely, this paper merges two complementary perspectives: an unsupervised analysis with PRISMA satellite imagery to explore whether these indices are truly distinct in practice and a supervised classification over UAV hyperspectral data. We assess their discriminative power, statistical correlations, and perceptual similarities. Our findings suggest that while many VIs have a certain correlation with the NDVI, meaningful differences emerge depending on landscape and application context, thus supporting their effectiveness as discriminative features usable in remote crop segmentation and recognition applications. Full article
(This article belongs to the Special Issue Selected Papers from International Symposium on Electronics and Telecommunications ISETC 2024)
►▼ Show Figures

Figure 1

16 pages, 1457 KiB  
Article
Benchmarking Accelerometer and CNN-Based Vision Systems for Sleep Posture Classification in Healthcare Applications
by Minh Long Hoang, Guido Matrella, Dalila Giannetto, Paolo Craparo and Paolo Ciampolini
Sensors 2025, 25(12), 3816; https://doi.org/10.3390/s25123816 - 18 Jun 2025
Viewed by 201
Abstract
Sleep position recognition plays a crucial role in diagnosing and managing various health conditions, such as sleep apnea, pressure ulcers, and musculoskeletal disorders. Accurate monitoring of body posture during sleep can provide valuable insights for clinicians and support the development of intelligent healthcare [...] Read more.
Sleep position recognition plays a crucial role in diagnosing and managing various health conditions, such as sleep apnea, pressure ulcers, and musculoskeletal disorders. Accurate monitoring of body posture during sleep can provide valuable insights for clinicians and support the development of intelligent healthcare systems. This research presents a comparative analysis of sleep position recognition using two distinct approaches: image-based deep learning and accelerometer-based classification. There are five classes: prone, supine, right side, left side, and wake up. For the image-based method, the Visual Geometry Group 16 (VGG16) convolutional neural network was fine-tuned with data augmentation strategies including rotation, reflection, scaling, and translation to enhance model generalization. The image-based model achieved an overall accuracy of 93.49%, with perfect precision and recall for “right side” and “wakeup” positions, but slightly lower performance for “left side” and “supine” classes. In contrast, the accelerometer-based method employed a feedforward neural network trained on features extracted from segmented accelerometer data, such as signal sum, standard deviation, maximum, and spike count. This method yielded superior performance, reaching an accuracy exceeding 99.8% across most sleep positions. The “wake up” position was particularly easy to detect due to the absence of body movements such as heartbeat or respiration when the person is no longer in bed. The results demonstrate that while image-based models are effective, accelerometer-based classification offers higher precision and robustness, particularly in real-time and privacy-sensitive scenarios. Further comparisons of the system characteristics, data size, and training time are also carried out to offer crucial insights for selecting the appropriate technology in clinical, in-home, or embedded healthcare monitoring applications. Full article
(This article belongs to the Special Issue Advances in Sensing Technologies for Sleep Monitoring)
►▼ Show Figures

Figure 1

34 pages, 18851 KiB  
Article
Dual-Branch Multi-Dimensional Attention Mechanism for Joint Facial Expression Detection and Classification
by Cheng Peng, Bohao Li, Kun Zou, Bowen Zhang, Genan Dai and Ah Chung Tsoi
Sensors 2025, 25(12), 3815; https://doi.org/10.3390/s25123815 - 18 Jun 2025
Viewed by 182
Abstract
This paper addresses the central issue arising from the (SDAC) of facial expressions, namely, to balance the competing demands of good global features for detection, and fine features for good facial expression classifications by replacing the feature extraction part of the “neck” network [...] Read more.
This paper addresses the central issue arising from the (SDAC) of facial expressions, namely, to balance the competing demands of good global features for detection, and fine features for good facial expression classifications by replacing the feature extraction part of the “neck” network in the feature pyramid network in the You Only Look Once X (YOLOX) framework with a novel architecture involving three attention mechanisms—batch, channel, and neighborhood—which respectively explores the three input dimensions—batch, channel, and spatial. Correlations across a batch of images in the individual path of the dual incoming paths are first extracted by a self attention mechanism in the batch dimension; these two paths are fused together to consolidate their information and then split again into two separate paths; the information along the channel dimension is extracted using a generalized form of channel attention, an adaptive graph channel attention, which provides each element of the incoming signal with a weight that is adapted to the incoming signal. The combination of these two paths, together with two skip connections from the input to the batch attention to the output of the adaptive channel attention, then passes into a residual network, with neighborhood attention to extract fine features in the spatial dimension. This novel dual path architecture has been shown experimentally to achieve a better balance between the competing demands in an SDAC problem than other competing approaches. Ablation studies enable the determination of the relative importance of these three attention mechanisms. Competitive results are obtained on two non-aligned face expression recognition datasets, RAF-DB and SFEW, when compared with other state-of-the-art methods. Full article
(This article belongs to the Special Issue Sensing Technologies Applied in Human Emotion and Facial Expression Recognition)
►▼ Show Figures

Figure 1

19 pages, 4970 KiB  
Article
LGFUNet: A Water Extraction Network in SAR Images Based on Multiscale Local Features with Global Information
by Xiaowei Bai, Yonghong Zhang and Jujie Wei
Sensors 2025, 25(12), 3814; https://doi.org/10.3390/s25123814 - 18 Jun 2025
Viewed by 159
Abstract
To address existing issues in water extraction from SAR images based on deep learning, such as confusion between mountain shadows and water bodies and difficulty in extracting complex boundary details for continuous water bodies, the LGFUNet model is proposed. The LGFUNet model consists [...] Read more.
To address existing issues in water extraction from SAR images based on deep learning, such as confusion between mountain shadows and water bodies and difficulty in extracting complex boundary details for continuous water bodies, the LGFUNet model is proposed. The LGFUNet model consists of three parts: the encoder–decoder, the DECASPP module, and the LGFF module. In the encoder–decoder, the Swin-Transformer module is used instead of convolution kernels for feature extraction, enhancing the learning of global information and improving the model’s ability to capture the spatial features of continuous water bodies. The DECASPP module is employed to extract and select multiscale features, focusing on complex water body boundary details. Additionally, a series of LGFF modules are inserted between the encoder and decoder to reduce the semantic gap between the encoder and decoder feature maps and the spatial information loss caused by the encoder’s downsampling process, improving the model’s ability to learn detailed information. Sentinel-1 SAR data from the Qinghai–Tibet Plateau region are selected, and the water extraction performance of the proposed LGFUNet model is compared with that of existing methods such as U-Net, Swin-UNet, and SCUNet++. The results show that the LGFUNet model achieves the best performance, respectively. Full article
(This article belongs to the Section Remote Sensors)
►▼ Show Figures

Figure 1

28 pages, 3951 KiB  
Article
An Iterative Error Correction Procedure for Single Sheet Testers Using FEM 3D Model
by Robert Krobot and Martin Dadić
Sensors 2025, 25(12), 3813; https://doi.org/10.3390/s25123813 - 18 Jun 2025
Viewed by 156
Abstract
Determination of single-valued BH curve and power loss curve of electric steels is an important parameter in the design of electrical machines and transformers. This paper proposes a correction procedure for the measurement of anhysteretic BH curve and power losses, based on the [...] Read more.
Determination of single-valued BH curve and power loss curve of electric steels is an important parameter in the design of electrical machines and transformers. This paper proposes a correction procedure for the measurement of anhysteretic BH curve and power losses, based on the finite element model (FEM) and SST apparatus. A 3D finite element model (FEM) of the SST (Single Sheet Tester) was developed with respect to the IEC 60404-3 standard. The measurement results obtained with a standardized SST apparatus are fed to its FEM and used to iteratively correct initial BH and power loss curves, obtained using magnetic equivalent circuits theory. The proposed iterative correction procedure is based on the steepest descent algorithm, while the stopping criteria were based on the difference between simulated and measured global variables (power loss, induced voltage, and primary current). After correction, root mean squared errors were decreased from 1.85 A/m to 42.88 × 10−3 A/m for the BH curve, and from 44.5 × 10−4 W/kg to 7.28 × 10−4 W/kg for the power loss curve. Full article
(This article belongs to the Special Issue Electronic Test and Measurement Instrumentation: Design and Applications (2nd Edition))
►▼ Show Figures

Figure 1

21 pages, 1523 KiB  
Article
Federated Learning for a Dynamic Edge: A Modular and Resilient Approach
by Leonardo Almeida, Rafael Teixeira, Gabriele Baldoni, Mário Antunes and Rui L. Aguiar
Sensors 2025, 25(12), 3812; https://doi.org/10.3390/s25123812 - 18 Jun 2025
Viewed by 293
Abstract
The increasing demand for distributed machine learning like Federated Learning (FL) in dynamic, resource-constrained edge environments, 5G/6G networks, and the proliferation of mobile and edge devices, presents significant challenges related to fault tolerance, elasticity, and communication efficiency. This paper addresses these issues by [...] Read more.
The increasing demand for distributed machine learning like Federated Learning (FL) in dynamic, resource-constrained edge environments, 5G/6G networks, and the proliferation of mobile and edge devices, presents significant challenges related to fault tolerance, elasticity, and communication efficiency. This paper addresses these issues by proposing a novel modular and resilient FL framework. In this context, resilience refers to the system’s ability to maintain operation and performance despite disruptions. The framework is built on decoupled modules handling core FL functionalities, allowing flexibility in integrating various algorithms, communication protocols, and resilience strategies. Results demonstrate the framework’s ability to integrate different communication protocols and FL paradigms, showing that protocol choice significantly impacts performance, particularly in high-volume communication scenarios, with Zenoh and MQTT exhibiting lower overhead than Kafka in tested configurations, and Zenoh emerging as the most efficient communication option. Additionally, the framework successfully maintained model training and achieved convergence even when simulating probabilistic worker failures, achieving a MCC of 0.9453. Full article
(This article belongs to the Special Issue Edge Computing in IoT Networks Based on Artificial Intelligence)
►▼ Show Figures

Figure 1

25 pages, 1733 KiB  
Article
Decentralized Communication-Free Controller for Synchronous Solar-Powered Water Pumping with Emulated Neighbor Sensing
by Roungsan Chaisricharoen, Wanus Srimaharaj, Punnarumol Temdee, Hamed Yahoui and Nina Bencheva
Sensors 2025, 25(12), 3811; https://doi.org/10.3390/s25123811 - 18 Jun 2025
Viewed by 141
Abstract
Solar-powered pumping systems using series pumps are commonly applied in the delivery of water to remote agricultural regions, particularly in hilly tropical terrain. The synchronization of these pumps typically depends on reliable communication; however, dense vegetation, elevation changes, and weather conditions often disrupt [...] Read more.
Solar-powered pumping systems using series pumps are commonly applied in the delivery of water to remote agricultural regions, particularly in hilly tropical terrain. The synchronization of these pumps typically depends on reliable communication; however, dense vegetation, elevation changes, and weather conditions often disrupt signals. To address these limitations, a fully decentralized, communication-free control system is proposed. Each pumping station operates independently while maintaining synchronized operation through emulated neighbor sensing. The system applies a discrete-time control algorithm with virtual sensing that estimates neighboring pump statuses. Each station consists of a solar photovoltaic (PV) array, variable-speed drive, variable inlet valve, reserve tank, and local control unit. The controller adjusts the valve positions and pump power based on real-time water level measurements and virtual neighbor sensing. The simulation results across four scenarios, including clear sky, cloudy conditions, temporary outage, and varied irradiance, demonstrated steady-state operation with no water overflow or shortage and a steady-state error less than 4% for 3 m3 transfer. The error decreased as the average power increased. The proposed method maintained system functionality under simulated power outage and variable irradiance, confirming its suitability for remote agricultural areas where communication infrastructure is limited. Full article
(This article belongs to the Special Issue Disruptive Technologies and Wireless Sensor Network Communication Algorithms (2nd Edition))
►▼ Show Figures

Figure 1

28 pages, 9320 KiB  
Article
Embedded Sensor Data Fusion and TinyML for Real-Time Remaining Useful Life Estimation of UAV Li Polymer Batteries
by Jutarut Chaoraingern and Arjin Numsomran
Sensors 2025, 25(12), 3810; https://doi.org/10.3390/s25123810 - 18 Jun 2025
Viewed by 207
Abstract
The accurate real-time estimation of the remaining useful life (RUL) of lithium-polymer (LiPo) batteries is a critical enabler for ensuring the safety, reliability, and operational efficiency of unmanned aerial vehicles (UAVs). Nevertheless, achieving such prognostics on resource-constrained embedded platforms remains a considerable technical [...] Read more.
The accurate real-time estimation of the remaining useful life (RUL) of lithium-polymer (LiPo) batteries is a critical enabler for ensuring the safety, reliability, and operational efficiency of unmanned aerial vehicles (UAVs). Nevertheless, achieving such prognostics on resource-constrained embedded platforms remains a considerable technical challenge. This study proposes an end-to-end TinyML-based framework that integrates embedded sensor data fusion with an optimized feedforward neural network (FFNN) model for efficient RUL estimation under strict hardware limitations. The system collects voltage, discharge time, and capacity measurements through a lightweight data fusion pipeline and leverages the Edge Impulse platform with the EON™Compiler for model optimization. The trained model is deployed on a dual-core ARM Cortex-M0+ Raspberry Pi RP2040 microcontroller, communicating wirelessly with a LabVIEW-based visualization system for real-time monitoring. Experimental validation on an 80-gram UAV equipped with a 1100 mAh LiPo battery demonstrates a mean absolute error (MAE) of 3.46 cycles and a root mean squared error (RMSE) of 3.75 cycles. Model testing results show an overall accuracy of 98.82%, with a mean squared error (MSE) of 55.68, a mean absolute error (MAE) of 5.38, and a variance score of 0.99, indicating strong regression precision and robustness. Furthermore, the quantized (int8) version of the model achieves an inference latency of 2 ms, with memory utilization of only 1.2 KB RAM and 11 KB flash, confirming its suitability for real-time deployment on resource-constrained embedded devices. Overall, the proposed framework effectively demonstrates the feasibility of combining embedded sensor data fusion and TinyML to enable accurate, low-latency, and resource-efficient real-time RUL estimation for UAV battery health management. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

31 pages, 14480 KiB  
Article
Vega: LLM-Driven Intelligent Chatbot Platform for Internet of Things Control and Development
by Harith Al-Safi, Harith Ibrahim and Paul Steenson
Sensors 2025, 25(12), 3809; https://doi.org/10.3390/s25123809 - 18 Jun 2025
Viewed by 340
Abstract
Large language models (LLMs) have revolutionized natural language processing (NLP), yet their potential in Internet of Things (IoT) and embedded systems (ESys) applications remains largely unexplored. Traditional IoT interfaces often require specialized knowledge, creating barriers for non-technical users. We present Vega, a modular [...] Read more.
Large language models (LLMs) have revolutionized natural language processing (NLP), yet their potential in Internet of Things (IoT) and embedded systems (ESys) applications remains largely unexplored. Traditional IoT interfaces often require specialized knowledge, creating barriers for non-technical users. We present Vega, a modular system that leverages LLMs to enable intuitive, natural language control and interrogation of IoT devices, specifically, a Raspberry Pi (RPi) connected to various sensors, actuators, and devices. Our solution comprises three key components: a physical circuit with input and output devices used to showcase the LLM’s ability to interact with hardware, an RPi integrating a control server, and a web application integrating LLM logic. Users interact with the system through natural language, which the LLM interprets to remotely call appropriate commands for the RPi. The RPi executes these instructions on the physically connected circuit, with outcomes communicated back to the user via LLM-generated responses. The system’s performance is empirically evaluated using a range of task complexities and user scenarios, demonstrating its ability to handle complex and conditional logic without additional coding on the RPi, reducing the need for extensive programming on IoT devices. We showcase the system’s real-world applicability through physical circuit implementation while providing insights into its limitations and potential scalability. Our findings reveal that LLM-driven IoT control can effectively bridge the gap between complex device functionality and user-friendly interaction, and also opens new avenues for creative and intelligent IoT applications. This research offers insights into the design and implementation of LLM-integrated IoT interfaces. Full article
(This article belongs to the Special Issue AI-Empowered Internet of Things)
►▼ Show Figures

Figure 1

17 pages, 10430 KiB  
Article
Intelligent Sports Weights
by Olga dos Santos Duarte, Gustavo Jacinto, Mário Véstias and Rui Policarpo Duarte
Sensors 2025, 25(12), 3808; https://doi.org/10.3390/s25123808 - 18 Jun 2025
Viewed by 194
Abstract
Weightlifting is a common fitness activity and can be practiced individually without supervision. However, performing regular weightlifting exercises without any form of feedback can lead to serious injuries. To counter this, this work proposes a different approach to automatic weightlifting supervision off-the-person. The [...] Read more.
Weightlifting is a common fitness activity and can be practiced individually without supervision. However, performing regular weightlifting exercises without any form of feedback can lead to serious injuries. To counter this, this work proposes a different approach to automatic weightlifting supervision off-the-person. The proposed embedded system is coupled to the weights and evaluates if they follow the correct trajectory in real time. The system is based on a low-power embedded System-on-a-Chip to perform the classification of the correctness of physical exercises using a Convolutional Neural Network with data from the embedded IMU. It is a low-cost solution and can be adapted to the characteristics of specific exercises to fine-tune the performance of the athlete. Experimental results show real-time monitoring capability with an average accuracy close to 95%. To favor its use, the prototypes have been enclosed on a custom 3D case and validated in an operational environment. All research outputs, developments, and engineering models are publicly available. Full article
(This article belongs to the Special Issue Edge AI for Wearables and IoT)
►▼ Show Figures

Figure 1

18 pages, 1032 KiB  
Article
AI for Sustainable Recycling: Efficient Model Optimization for Waste Classification Systems
by Oriol Chacón-Albero, Mario Campos-Mocholí, Cédric Marco-Detchart, Vicente Julian, Jaime Andrés Rincon and Vicent Botti
Sensors 2025, 25(12), 3807; https://doi.org/10.3390/s25123807 - 18 Jun 2025
Viewed by 337
Abstract
The increasing volume of global waste presents a critical environmental and societal challenge, demanding innovative solutions to support sustainable practices such as recycling. Advances in Computer Vision (CV) have enabled automated waste recognition systems that guide users in correctly sorting their waste, with [...] Read more.
The increasing volume of global waste presents a critical environmental and societal challenge, demanding innovative solutions to support sustainable practices such as recycling. Advances in Computer Vision (CV) have enabled automated waste recognition systems that guide users in correctly sorting their waste, with state-of-the-art architectures achieving high accuracy. More recently, attention has shifted toward lightweight and efficient models suitable for mobile and edge deployment. These systems process data from integrated camera sensors in Internet of Things (IoT) devices, operating in real time to classify waste at the point of disposal, whether embedded in smart bins, mobile applications, or assistive tools for household use. In this work, we extend our previous research by improving both dataset diversity and model efficiency. We introduce an expanded dataset that includes an organic waste class and more heterogeneous images, and evaluate a range of quantized CNN models to reduce inference time and resource usage. Additionally, we explore ensemble strategies using aggregation functions to boost classification performance, and validate selected models on real embedded hardware and under simulated lighting variations. Our results support the development of robust, real-time recycling assistants for resource-constrained devices. We also propose architectural deployment scenarios for smart containers, and cloud-assisted solutions. By improving waste sorting accuracy, these systems can help reduce landfill use, support citizen engagement through real-time feedback, increase material recovery, support data-informed environmental decision making, and ease operational challenges for recycling facilities caused by misclassified materials. Ultimately, this contributes to circular economy objectives and advances the broader field of environmental intelligence. Full article
(This article belongs to the Special Issue Internet of Things (IoT) and Cloud Computing Technologies for Smart Cities and Rural Areas: Sensing and Information Management Technologies)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-264
Previous Issue
Next Issue
Back to TopTop