Next Issue
Volume 25, May-2
Previous Issue
Volume 25, April-2
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
7.3
3.4
Journals
Sensors
Volume 25
Issue 9
sensors-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Sensors, Volume 25, Issue 9 (May-1 2025) – 322 articles

Cover Story (view full-size image): This study introduces ion gel-modulated IGZO phototransistors fabricated via inkjet printing, achieving breakthrough synaptic emulation across 450–638 nm wavelengths. The ≤150 °C fabrication process ensures compatibility with future flexible substrates while preserving the exceptional electrical performance of IGZO. By engineering oxygen vacancy-induced subgap states, the devices enable red-light (638 nm) synaptic plasticity through multi-step photon absorption. The coplanar ion gel gate architecture delivers ultrahigh specific capacitance (2.78 μF/cm2), supporting ultralow-voltage operation with biological fidelity in short- and long-term plasticity, as well as in paired-pulse facilitation. This platform bridges optoelectronic signal processing with neuromorphic learning, offering scalable manufacturing routes for artificial vision systems. 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:
13 pages, 1278 KiB  
Article
Copper Phthalocyanine Chemiresistors as Industrial NO2 Alarms
by Hadi AlQahtani, Mohammad Alshammari, Amjad M. Kamal and Martin Grell
Sensors 2025, 25(9), 2955; https://doi.org/10.3390/s25092955 - 7 May 2025
Viewed by 194
Abstract
We present a chemiresistor sensor for NO2 leaks. The sensor uses the organometallic semiconductor copper(II)phthalocyanine (CuPc), and is more easily manufactured and characterised than previously described organic transistor gas sensors. Resistance R is high but within the range of modern voltage buffers. [...] Read more.
We present a chemiresistor sensor for NO2 leaks. The sensor uses the organometallic semiconductor copper(II)phthalocyanine (CuPc), and is more easily manufactured and characterised than previously described organic transistor gas sensors. Resistance R is high but within the range of modern voltage buffers. The chemiresistor weakly responds to several gases, with either a small increase (NH3 and H2S) or decrease (SO2) in R. However, the response is low at environmental pollution levels. The response to NO2 also is near-zero for permitted long-term exposure. Our sensor is, therefore, not suited for environmental monitoring, but acceptable environmental pollutant levels do not interfere with the sensor. Above a threshold of ~87 ppb, the response to NO2 becomes very strong. This response is presumably due to the doping of CuPc by the strongly oxidising NO2, and is far stronger than for previously reported CuPc chemiresistors. We relate this to differences in the film morphology. Under 1 ppm NO2, R drops by a factor of 870 vs. non-polluted air. An amount of 1 ppm NO2 is far above the ‘background’ environmental pollution, thereby avoiding false alarms, but far below immediately life-threatening levels, thus giving time to evacuate. Our sensor is destined for leak detection in the nitrogen fertiliser industry, where NO2 is an important intermediate. Full article
(This article belongs to the Section Industrial Sensors)
►▼ Show Figures

Graphical abstract

17 pages, 4101 KiB  
Article
Dynamic Parameterization and Optimized Flight Paths for Enhanced Aeromagnetic Compensation in Large Unmanned Aerial Vehicles
by Zhentao Yu, Liwei Ye, Can Ding, Cheng Chi, Cong Liu and Pu Cheng
Sensors 2025, 25(9), 2954; https://doi.org/10.3390/s25092954 - 7 May 2025
Viewed by 199
Abstract
Aeromagnetic detection is a geophysical exploration technology that utilizes aircraft-mounted magnetometers to map variations in the Earth’s magnetic field. As a critical methodology for subsurface investigations, it has been extensively applied in geological mapping, mineral resource prospecting, hydrocarbon exploration, and engineering geological assessments. [...] Read more.
Aeromagnetic detection is a geophysical exploration technology that utilizes aircraft-mounted magnetometers to map variations in the Earth’s magnetic field. As a critical methodology for subsurface investigations, it has been extensively applied in geological mapping, mineral resource prospecting, hydrocarbon exploration, and engineering geological assessments. However, the metallic composition of aircraft platforms inherently generates magnetic interference, which significantly distorts the measurements acquired by onboard magnetometers. Aeromagnetic compensation aims to mitigate these platform-induced magnetic disturbances, thereby enhancing the accuracy of magnetic anomaly detection. Building upon the conventional Tolles-Lawson (T-L) model, this study introduces an enhanced compensation framework that addresses two key limitations: (1) minor deformations that occur due to the non-rigidity of the aircraft fuselage, resulting in additional interfering magnetic fields, and (2) coupled interference between geomagnetic field variations and aircraft maneuvers. The proposed model expands the original 18 compensation coefficients to 57 through dynamic parameterization, achieving a 22.41% improvement in compensation efficacy compared with the traditional T-L model. Furthermore, recognizing the operational challenges of large unmanned aerial vehicles (UAVs) in conventional calibration flights, this work redesigns the flight protocol by eliminating high-risk yaw maneuvers and optimizing the flight path geometry. Experimental validations conducted in the South China Sea demonstrate exceptional performance, with the interference magnetic field reduced to 0.0385 nT (standard deviation) during level flight, achieving an improvement ratio (IR) of 4.1688. The refined methodology not only enhances compensation precision but also substantially improves operational safety for large UAVs, offering a robust solution for modern aeromagnetic surveys. Full article
(This article belongs to the Section Navigation and Positioning)
►▼ Show Figures

Figure 1

28 pages, 8922 KiB  
Article
Multi-Robot Cooperative Simultaneous Localization and Mapping Algorithm Based on Sub-Graph Partitioning
by Wan Xu, Yanliang Chen, Shijie Liu, Ao Nie and Rupeng Chen
Sensors 2025, 25(9), 2953; https://doi.org/10.3390/s25092953 - 7 May 2025
Viewed by 215
Abstract
To address the challenges in multi-robot collaborative SLAM, including excessive redundant computations and low processing efficiency in candidate loop closure selection during front-end loop detection, as well as high computational complexity and long iteration times due to global pose optimization in the back-end, [...] Read more.
To address the challenges in multi-robot collaborative SLAM, including excessive redundant computations and low processing efficiency in candidate loop closure selection during front-end loop detection, as well as high computational complexity and long iteration times due to global pose optimization in the back-end, this paper introduces several key improvements. First, a global matching and candidate loop selection strategy is incorporated into the front-end loop detection module, leveraging both LiDAR point clouds and visual features to achieve cross-robot loop detection, effectively mitigating computational redundancy and reducing false matches in collaborative multi-robot systems. Second, an improved distributed robust pose graph optimization algorithm is proposed in the back-end module. By introducing a robust cost function to filter out erroneous loop closures and employing a subgraph optimization strategy during iterative optimization, the proposed approach enhances convergence speed and solution quality, thereby reducing uncertainty in multi-robot pose association. Experimental results demonstrate that the proposed method significantly improves computational efficiency and localization accuracy. Specifically, in front-end loop detection, the proposed algorithm achieves an F1-score improvement of approximately 8.5–51.5% compared to other methods. In back-end optimization, it outperforms traditional algorithms in terms of both convergence speed and optimization accuracy. In terms of localization accuracy, the proposed method achieves an improvement of approximately 32.8% over other open source algorithms. Full article
(This article belongs to the Section Sensors and Robotics)
►▼ Show Figures

Figure 1

37 pages, 3827 KiB  
Review
Research Progress on Data-Driven Industrial Fault Diagnosis Methods
by Liang Lei, Weibin Li, Shiwei Zhang, Changyuan Wu and Hongxiang Yu
Sensors 2025, 25(9), 2952; https://doi.org/10.3390/s25092952 - 7 May 2025
Viewed by 255
Abstract
With the advent of Industry 5.0, fault diagnosis is playing an increasingly important role in routine equipment maintenance and condition monitoring. From the perspective of industrial big data, this paper systematically reviews the current mainstream industrial fault diagnosis methods. The content covers the [...] Read more.
With the advent of Industry 5.0, fault diagnosis is playing an increasingly important role in routine equipment maintenance and condition monitoring. From the perspective of industrial big data, this paper systematically reviews the current mainstream industrial fault diagnosis methods. The content covers the main sources of industrial big data, commonly used datasets, and the construction of related platforms. In conjunction with the development of multi-source heterogeneous data, the paper explores the evolutionary path of fault diagnosis methods. Subsequently, it provides an in-depth analysis of data-driven fault diagnosis techniques in industrial applications, with particular emphasis on the pivotal role of deep learning algorithms in fault diagnosis. Next, it discusses the applications and development of large models in the field of fault diagnosis, focusing on their potential to enhance diagnostic intelligence and generalization under big data environments. Finally, the paper looks ahead to the future development of data-driven fault diagnosis methods, pointing out that data quality, interpretability of deep learning, and edge-based large models are important research directions that urgently require breakthroughs. Full article
(This article belongs to the Section Fault Diagnosis & Sensors)
►▼ Show Figures

Figure 1

22 pages, 3546 KiB  
Article
Nano-Tailored Triple Gas Sensor for Real-Time Monitoring of Dough Preparation in Kitchen Machines
by Dario Genzardi, Immacolata Caruso, Elisabetta Poeta, Veronica Sberveglieri and Estefanía Núñez Carmona
Sensors 2025, 25(9), 2951; https://doi.org/10.3390/s25092951 - 7 May 2025
Viewed by 172
Abstract
We evaluated the efficacy of an innovative technique using an S3+ device equipped with two custom-made nanosensors (e-nose). These sensors are integrated into kitchen appliances, such as planetary mixers, to monitor and assess dough leavening from preparation to the fully risen stage. Since [...] Read more.
We evaluated the efficacy of an innovative technique using an S3+ device equipped with two custom-made nanosensors (e-nose). These sensors are integrated into kitchen appliances, such as planetary mixers, to monitor and assess dough leavening from preparation to the fully risen stage. Since monitoring in domestic appliances is often subjective and non-reproducible, this approach aims to ensure safe, high-quality, and consistent results for consumers. Two sensor chips, each with three metal oxide semiconductor (MOS) elements, were used to assess doughs prepared with flours of varying strengths (W200, W250, W390). Analyses were conducted continuously (from the end of mixing to 1.5 h of leavening) and in two distinct phases: pre-leavening (PRE) and post-leavening (POST). The technique was validated through solid-phase micro-extraction combined with gas chromatography–mass spectrometry (SPME-GC-MS), used to analyze volatile profiles in both phases. The S3+ device clearly discriminated between PRE and POST samples in 3D Linear Discriminant Analysis (LDA) plots, while 2D LDA confirmed flour-type discrimination during continuous leavening. These findings were supported by SPME-GC-MS results, highlighting differences in the volatile organic compound (VOC) profiles. The system achieved 100% classification accuracy between PRE and POST stages and effectively distinguished all flour types. Integrating this e-nose into kitchen equipment offers a concrete opportunity to optimize leavening by identifying the ideal endpoint, improving reproducibility, and reducing waste. In future applications, sensor data could support feedback control systems capable of adjusting fermentation parameters like time and temperature in real time. Full article
(This article belongs to the Section Chemical Sensors)
►▼ Show Figures

Figure 1

23 pages, 5085 KiB  
Article
Analysis of Anti-Jamming Performance of HF Access Network Based on Asymmetric Frequency Hopping
by Ruijie Duan, Liang Jin and Xiaofei Lan
Sensors 2025, 25(9), 2950; https://doi.org/10.3390/s25092950 - 7 May 2025
Viewed by 171
Abstract
The primary focus of this paper lies in addressing the inadequate anti-dynamic jamming capability of the link layer within high-frequency (HF) access networks. To this end, we propose the incorporation of asymmetric frequency-hopping (AFH) technology within the wireless communication segment of HF access [...] Read more.
The primary focus of this paper lies in addressing the inadequate anti-dynamic jamming capability of the link layer within high-frequency (HF) access networks. To this end, we propose the incorporation of asymmetric frequency-hopping (AFH) technology within the wireless communication segment of HF access networks. This innovation aims to supersede the existing fixed-frequency and frequency-hopping communication methodologies, ultimately enhancing the network’s resilience against dynamic jamming. Moreover, we undertake a modeling analysis to delve into the ramifications of asymmetric frequency-hopping communication in dynamic jamming environments. This modeling framework serves to elucidate the dynamics of user spectrum occupation and jamming occurrences. Our proposed methodology leverages a two-dimensional Markov queuing model, equipped with a single server, for the purpose of managing the spectrum allocation within HF access network subnets. Consequently, the base station gains the capability to dynamically manage and adjust the available spectrum in real time, thereby effectively mitigating mutual jamming among users and facilitating the seamless implementation of asymmetric frequency hopping in HF access networks. Lastly, we conduct a simulation analysis to evaluate the changes in anti-jamming performance indices within the HF access network. This analysis compares the merits and demerits of utilizing fixed-frequency, frequency-hopping, and asymmetric frequency-hopping communication techniques. Our findings conclusively demonstrate that the integration of asymmetric frequency-hopping technology can significantly reduce outage and mutual jamming rates within HF access network subnets, thereby substantially bolstering their anti-jamming prowess. Full article
(This article belongs to the Topic Advances in Wireless and Mobile Networking)
►▼ Show Figures

Figure 1

18 pages, 5099 KiB  
Article
Surface Electromyographic Features for Severity Classification in Facial Palsy: Insights from a German Cohort and Implications for Future Biofeedback Use
by Ibrahim Manzoor, Aryana Popescu, Alexia Stark, Mykola Gorbachuk, Aldo Spolaore, Marcos Tatagiba, Georgios Naros and Kathrin Machetanz
Sensors 2025, 25(9), 2949; https://doi.org/10.3390/s25092949 - 7 May 2025
Viewed by 205
Abstract
Facial palsy (FP) significantly impacts patients’ quality of life. The accurate classification of FP severity is crucial for personalized treatment planning. Additionally, electromyographic (EMG)-based biofeedback shows promising results in improving recovery outcomes. This prospective study aims to identify EMG time series features that [...] Read more.
Facial palsy (FP) significantly impacts patients’ quality of life. The accurate classification of FP severity is crucial for personalized treatment planning. Additionally, electromyographic (EMG)-based biofeedback shows promising results in improving recovery outcomes. This prospective study aims to identify EMG time series features that can both classify FP and facilitate biofeedback. Therefore, it investigated surface EMG in FP patients and healthy controls during three different facial movements. Repeated-measures ANOVAs (rmANOVA) were conducted to examine the effects of MOTION (move/rest), SIDE (healthy/lesioned) and the House–Brackmann score (HB), across 20 distinct EMG parameters. Correlation analysis was performed between HB and the asymmetry index of EMG parameters, complemented by Fisher score calculations to assess feature relevance in distinguishing between HB levels. Overall, 55 subjects (51.2 ± 14.73 years, 35 female) were included in the study. RmANOVAs revealed a highly significant effect of MOTION across almost all movement types (p < 0.001). Integrating the findings from rmANOVA, the correlation analysis and Fisher score analysis, at least 5/20 EMG parameters were determined to be robust indicators for assessing the degree of paresis and guiding biofeedback. This study demonstrates that EMG can reliably determine severity and guide effective biofeedback in FP, and in severe cases. Our findings support the integration of EMG into personalized rehabilitation strategies. However, further studies are mandatory to improve recovery outcomes. Full article
(This article belongs to the Special Issue Motion Control Using EMG Signals)
►▼ Show Figures

Figure 1

18 pages, 613 KiB  
Article
Covert Communication Scheme for OOK in Asymmetric Noise Systems
by Weicheng Xu, Xiaopeng Ji and Ruizhi Zhu
Sensors 2025, 25(9), 2948; https://doi.org/10.3390/s25092948 - 7 May 2025
Viewed by 99
Abstract
Existing covert communication schemes based on On–Off Keying (OOK) have not considered asymmetric noise environments, which limits their applicability in complex communication scenarios such as terahertz and underwater acoustic covert communications. To address this issue, this paper proposes a phase-based OOK coding scheme. [...] Read more.
Existing covert communication schemes based on On–Off Keying (OOK) have not considered asymmetric noise environments, which limits their applicability in complex communication scenarios such as terahertz and underwater acoustic covert communications. To address this issue, this paper proposes a phase-based OOK coding scheme. In particular, the transmitter Alice can adjust the initial phase of the transmitted symbol to align the signal with the stronger noise components in asymmetric noise communication scenarios, thereby exploiting the masking effect of noise to achieve covert transmission. To quantify performance, the KL divergence and mutual information of the OOK coding scheme are adopted as measures of covertness and transmission performance, respectively. An optimization problem involving the input signal distribution an, signal amplitude β, and initial phase angle θ is formulated and solved to obtain the maximum covert transmission rate. Numerical results demonstrate that in asymmetric noise systems, the initial phase angle and the Gaussian noise components on the real and imaginary axes of the complex plane influence both covertness performance and transmission rate. Adjusting the initial phase towards the direction with lower noise components can maximally suppress noise interference, thereby improving the covertness performance. Full article
(This article belongs to the Section Communications)
►▼ Show Figures

Figure 1

21 pages, 4513 KiB  
Article
An Enhanced ZigBee-Based Indoor Localization Method Using Multi-Stage RSSI Filtering and LQI-Aware MLE
by Jianming Li, Shuyan Yu, Zhe Wei and Zhanpeng Zhou
Sensors 2025, 25(9), 2947; https://doi.org/10.3390/s25092947 - 7 May 2025
Viewed by 192
Abstract
Accurate indoor localization in wireless sensor networks remains a non-trivial challenge, particularly in complex environments characterized by signal variability and multipath propagation. This study presents a ZigBee-based localization approach that integrates multi-stage preprocessing of received signal strength indicator (RSSI) data with a reliability-aware [...] Read more.
Accurate indoor localization in wireless sensor networks remains a non-trivial challenge, particularly in complex environments characterized by signal variability and multipath propagation. This study presents a ZigBee-based localization approach that integrates multi-stage preprocessing of received signal strength indicator (RSSI) data with a reliability-aware extension of the maximum likelihood estimation (MLE) algorithm. To improve measurement stability, a hybrid filtering framework combining Kalman filtering, Dixon’s Q test, Gaussian smoothing, and mean averaging is applied to reduce the influence of noise and outliers. Building on the filtered data, the proposed method introduces a noise and link quality indicator (LQI)-based dynamic weighting mechanism that adjusts the contribution of each distance estimate during localization. The approach was evaluated under simulated and semi-physical non-line-of-sight (NLOS) indoor conditions designed to reflect practical deployment scenarios. While based on a limited set of representative test points, the method yielded improved positioning consistency and achieved an average accuracy gain of 11.7% over conventional MLE in the tested environments. These results suggest that the proposed method may offer a feasible solution for resource-constrained localization applications requiring robustness to signal degradation. Full article
(This article belongs to the Special Issue Sensors and Techniques for Indoor Positioning and Localization: 2nd Edition)
►▼ Show Figures

Figure 1

28 pages, 11967 KiB  
Article
Study on Spark Image Detection for Abrasive Belt Grinding via Transfer Learning with YOLOv8
by Jian Huang and Guangpeng Zhang
Sensors 2025, 25(9), 2946; https://doi.org/10.3390/s25092946 - 7 May 2025
Viewed by 189
Abstract
Aiming to solve the problems of low precision and poor efficiency caused by relying on manual experience during the manual polishing of blades, a multi-view spark image detection method based on YOLOv8 transfer learning is proposed. A multi-pose spark image dataset including front, [...] Read more.
Aiming to solve the problems of low precision and poor efficiency caused by relying on manual experience during the manual polishing of blades, a multi-view spark image detection method based on YOLOv8 transfer learning is proposed. A multi-pose spark image dataset including front, side, and 45° angle views is constructed, and the cross-view detection task is achieved for the first time. The generalization ability of the model is enhanced through the following innovative strategies: (1) a cross-view transfer learning framework based on dynamic anchor box optimization is designed, and the parameters of the front spark detection model YOLOv8 are transferred to the side and 45°-angle detection tasks; (2) an attention-guided feature alignment module is introduced to alleviate the feature distribution shift caused by view differences; and (3) a curriculum learning strategy is adopted, where the datasets of different views are trained separately first and then sampled to reconstruct the dataset for further training, gradually increasing the weight of samples from complex views. The experimental results show that on the self-built multi-view dataset (containing 3000 annotated images), this method achieves an average detection accuracy of 98.7%, which is 14.2% higher than that of the original YOLOv8 model. The inference speed reaches 55 FPS on an NVIDIA RTX 4090, meeting the requirements of industrial online monitoring. The research results provide key technical support for the intelligent prediction of the material removal rate in the precision machining of blades and have the potential for rapid deployment in industrial scenarios. Full article
(This article belongs to the Section Sensing and Imaging)
►▼ Show Figures

Figure 1

14 pages, 2429 KiB  
Article
End-to-End Architecture for Real-Time IoT Analytics and Predictive Maintenance Using Stream Processing and ML Pipelines
by Ouiam Khattach, Omar Moussaoui and Mohammed Hassine
Sensors 2025, 25(9), 2945; https://doi.org/10.3390/s25092945 - 7 May 2025
Viewed by 369
Abstract
The rapid proliferation of Internet of Things (IoT) devices across industries has created a need for robust, scalable, and real-time data processing architectures capable of supporting intelligent analytics and predictive maintenance. This paper presents a novel comprehensive architecture that enables end-to-end processing of [...] Read more.
The rapid proliferation of Internet of Things (IoT) devices across industries has created a need for robust, scalable, and real-time data processing architectures capable of supporting intelligent analytics and predictive maintenance. This paper presents a novel comprehensive architecture that enables end-to-end processing of IoT data streams, from acquisition to actionable insights. The system integrates Kafka-based message brokering for the high-throughput ingestion of real-time sensor data, with Apache Spark facilitating batch and stream extraction, transformation, and loading (ETL) processes. A modular machine-learning pipeline handles automated data preprocessing, training, and evaluation across various models. The architecture incorporates continuous monitoring and optimization components to track system performance and model accuracy, feeding insights to users via a dedicated Application Programming Interface (API). The design ensures scalability, flexibility, and real-time responsiveness, making it well suited for industrial IoT applications requiring continuous monitoring and intelligent decision-making. Full article
(This article belongs to the Special Issue IoT Sensor Systems: Design, Interfaces, Signals, Processing, and Applications)
►▼ Show Figures

Figure 1

20 pages, 3769 KiB  
Article
Comparative Analysis of Machine Learning Approaches for Fetal Movement Detection with Linear Acceleration and Angular Rate Signals
by Lucy Spicher, Carrie Bell, Kathleen H. Sienko and Xun Huan
Sensors 2025, 25(9), 2944; https://doi.org/10.3390/s25092944 - 7 May 2025
Viewed by 224
Abstract
Reduced fetal movement (RFM) can indicate that a fetus is at risk, but current monitoring methods provide only a “snapshot in time” of fetal health and require trained clinicians in clinical settings. To improve antenatal care, there is a need for continuous, objective [...] Read more.
Reduced fetal movement (RFM) can indicate that a fetus is at risk, but current monitoring methods provide only a “snapshot in time” of fetal health and require trained clinicians in clinical settings. To improve antenatal care, there is a need for continuous, objective fetal movement monitoring systems. Wearable sensors, like inertial measurement units (IMUs), offer a promising data-driven solution, but distinguishing fetal movements from maternal movements remains challenging. The potential benefits of using linear acceleration and angular rate data for fetal movement detection have not been fully explored. In this study, machine learning models were developed using linear acceleration and angular rate data from twenty-three participants who wore four abdominal IMUs and one chest reference while indicating perceived fetal movements with a handheld button. Random forest (RF), bi-directional long short-term memory (BiLSTM), and convolutional neural network (CNN) models were trained using hand-engineered features, time series data, and time–frequency spectrograms, respectively. The results showed that combining accelerometer and gyroscope data improved detection performance across all models compared to either one alone. CNN consistently outperformed other models but required larger datasets. RF and BiLSTM, while more sensitive to signal noise, offered reasonable performance with smaller datasets and greater interpretability. Full article
(This article belongs to the Special Issue Wearable Sensors for Continuous Health Monitoring and Analysis)
►▼ Show Figures

Graphical abstract

15 pages, 8743 KiB  
Article
Inverse Synthetic Aperture Radar Sparse Imaging Recovery Technique Based on Improved Alternating Direction Method of Multipliers
by Hongxing Hao, Wenjie Zhu, Ronghuan Yu and Desheng Liu
Sensors 2025, 25(9), 2943; https://doi.org/10.3390/s25092943 - 7 May 2025
Viewed by 143
Abstract
Inverse synthetic aperture radar (ISAR) technology is widely used in the field of target recognition. This research addresses the image reconstruction error in sparse imaging for bistatic radar systems. In this paper, sparse imaging technology is studied, and a sparse imaging recovery algorithm [...] Read more.
Inverse synthetic aperture radar (ISAR) technology is widely used in the field of target recognition. This research addresses the image reconstruction error in sparse imaging for bistatic radar systems. In this paper, sparse imaging technology is studied, and a sparse imaging recovery algorithm based on an improved Alternating Direction Method of Multipliers is proposed. The algorithm accelerates the convergence of the algorithm by dynamically adjusting iterative parameters in the iterative process. Experiments show that the algorithm proposed in this paper has lower relative recovery error in the case of different noise levels and sparsity, and it can be concluded that the algorithm proposed in this paper has a lower relative recovery error than the ADMMs (Alternating Direction Method of Multipliers). Full article
(This article belongs to the Section Radar Sensors)
►▼ Show Figures

Figure 1

12 pages, 1734 KiB  
Article
From Accelerometer to Cognition: Hand Motion Can Reflect Effects of Cardiac Coherence on Cognitive Flexibility
by Alix Bouni, Laurent M. Arsac, Olivier Chevalerias and Véronique Deschodt-Arsac
Sensors 2025, 25(9), 2942; https://doi.org/10.3390/s25092942 - 7 May 2025
Viewed by 143
Abstract
Hand displacements during task-directed movements are not random, but exhibit fractal behavior. Wearable sensing e.g., accelerometer-derived hand movement fluctuations, could add a significant contribution to cognitive and behavioral sciences, by accounting for fractal dynamics. In particular, multifractal testing of fluctuation time series has [...] Read more.
Hand displacements during task-directed movements are not random, but exhibit fractal behavior. Wearable sensing e.g., accelerometer-derived hand movement fluctuations, could add a significant contribution to cognitive and behavioral sciences, by accounting for fractal dynamics. In particular, multifractal testing of fluctuation time series has been shown to reflect the adaptive use of cognition, i.e., cognitive flexibility. This important property might be enhanced by an improved mental state. Here, an experimental group (16 participants, 3 females) practiced 5 min cardiac coherence (CC) prior to a cognitive flexibility task and was compared to a control group (13 participants, 4 females). Accelerometer-derived hand motion was analyzed using multifractal-multiscale detrended fluctuation analysis (MFMS-DFA) during a task involving cognitive flexibility, the Wisconsin Card Sorting Test (WCST). WCST included four phases alternating the use of cards with original shapes or animal pictures developed for children in previous research. Hand behavioral time series derived from the wearable accelerometer effectively exhibited nonlinear multifractality as shown using linearized surrogates testing. Multifractal-multiscale metrics revealed significant effects of pre-task CC practice, specifically during WCST shape condition where CC participants showed lower multifractal degree despite similar performances (perseverative errors). We conclude that multifractal-multiscale testing of accelerometer-derived hand motion could make a significant contribution to interpreting changes in cognitive flexibility. Full article
(This article belongs to the Special Issue Wearable Sensing Technology for Physiological and Behavioral Human Monitoring—2nd Edition)
►▼ Show Figures

Figure 1

24 pages, 1962 KiB  
Article
Multi-Variable Transformer-Based Meta-Learning for Few-Shot Fault Diagnosis of Large-Scale Systems
by Weiyang Li, Yixin Nie and Fan Yang
Sensors 2025, 25(9), 2941; https://doi.org/10.3390/s25092941 - 7 May 2025
Viewed by 177
Abstract
Fault diagnosis in large-scale systems presents significant challenges due to the complexity and high dimensionality of data, as well as the scarcity of labeled fault data, which are hard to obtain during the practical operation process. This paper proposes a novel approach, called [...] Read more.
Fault diagnosis in large-scale systems presents significant challenges due to the complexity and high dimensionality of data, as well as the scarcity of labeled fault data, which are hard to obtain during the practical operation process. This paper proposes a novel approach, called Multi-Variable Meta-Transformer (MVMT), to tackle these challenges. In order to deal with the multi-variable time series data, we modify the Transformer model, which is the currently most popular model on feature extraction of time series. To enable the Transformer model to simultaneously receive continuous and state inputs, we introduced feature layers before the encoder to better integrate the characteristics of both continuous and state variables. Then, we adopt the modified model as the base model for meta-learning—more specifically, the Model-Agnostic Meta-Learning (MAML) strategy. The proposed method leverages the power of Transformers for handling multi-variable time series data and employs meta-learning to enable few-shot learning capabilities. The case studies conducted on the Tennessee Eastman Process database and a Power-Supply System database demonstrate the exceptional performance of fault diagnosis in few-shot scenarios, whether based on continuous-only data or a combination of continuous and state variables. Full article
(This article belongs to the Special Issue Advances in Sensors for Online Condition Monitoring and Fault Diagnosis)
►▼ Show Figures

Figure 1

21 pages, 52785 KiB  
Article
MC-ASFF-ShipYOLO: Improved Algorithm for Small-Target and Multi-Scale Ship Detection for Synthetic Aperture Radar (SAR) Images
by Yubin Xu, Haiyan Pan, Lingqun Wang and Ran Zou
Sensors 2025, 25(9), 2940; https://doi.org/10.3390/s25092940 - 7 May 2025
Viewed by 231
Abstract
Synthetic aperture radar (SAR) ship detection holds significant application value in maritime monitoring, marine traffic management, and safety maintenance. Despite remarkable advances in deep-learning-based detection methods, performance remains constrained by the vast size differences between ships, limited feature information of small targets, and [...] Read more.
Synthetic aperture radar (SAR) ship detection holds significant application value in maritime monitoring, marine traffic management, and safety maintenance. Despite remarkable advances in deep-learning-based detection methods, performance remains constrained by the vast size differences between ships, limited feature information of small targets, and complex environmental interference in SAR imagery. Although many studies have separately tackled small target identification and multi-scale detection in SAR imagery, integrated approaches that jointly address both challenges within a unified framework for SAR ship detection are still relatively scarce. This study presents MC-ASFF-ShipYOLO (Monte Carlo Attention—Adaptively Spatial Feature Fusion—ShipYOLO), a novel framework addressing both small target recognition and multi-scale ship detection challenges. Two key innovations distinguish our approach: (1) We introduce a Monte Carlo Attention (MCAttn) module into the backbone network that employs random sampling pooling operations to generate attention maps for feature map weighting, enhancing focus on small targets and improving their detection performance. (2) We add Adaptively Spatial Feature Fusion (ASFF) modules to the detection head that adaptively learn spatial fusion weights across feature layers and perform dynamic feature fusion, ensuring consistent ship representations across scales and mitigating feature conflicts, thereby enhancing multi-scale detection capability. Experiments are conducted on a newly constructed dataset combining HRSID and SSDD. Ablation experiment results demonstrate that, compared to the baseline, MC-ASFF-ShipYOLO achieves improvements of 1.39% in precision, 2.63% in recall, 2.28% in AP50, and 3.04% in AP, indicating a significant enhancement in overall detection performance. Furthermore, comparative experiments show that our method outperforms mainstream models. Even under high-confidence thresholds, MC-ASFF-ShipYOLO is capable of predicting more high-quality detection boxes, offering a valuable solution for advancing SAR ship detection technology. Full article
(This article belongs to the Special Issue Recent Advances in Synthetic Aperture Radar (SAR) Remote Sensing)
►▼ Show Figures

Figure 1

17 pages, 4243 KiB  
Article
Estimation of Respiratory States Based on a Measurement Model of Airflow Characteristics in Powered Air-Purifying Respirators Using Differential Pressure and Pulse Width Modulation Control Signals—In the Development of a Public-Oriented Powered Air-Purifying Respirator as an Alternative to Lockdown Measures
by Yusaku Fujii, Akihiro Takita, Seiji Hashimoto and Kenji Amagai
Sensors 2025, 25(9), 2939; https://doi.org/10.3390/s25092939 - 7 May 2025
Viewed by 205
Abstract
Fluid dynamics modeling was conducted for the supply unit of a Powered Air-Purifying Respirator (PAPR) consisting of a nonwoven fabric filter and a pump, as well as for the exhaust filter (nonwoven fabric). The supply flow rate Q1 was modeled as a [...] Read more.
Fluid dynamics modeling was conducted for the supply unit of a Powered Air-Purifying Respirator (PAPR) consisting of a nonwoven fabric filter and a pump, as well as for the exhaust filter (nonwoven fabric). The supply flow rate Q1 was modeled as a function of the differential pressure ΔP and the duty value d of the PWM control under a constant pump voltage of V = 12.0 [V]. In contrast, the exhaust flow rate Q2 was modeled solely as a function of ΔP. To simulate the pressurized hood compartment of the PAPR, a pressure buffer and a connected “respiratory airflow simulator” (a piston–cylinder mechanism) were developed. The supply unit and exhaust filter were connected to this pressure buffer, and simulated respiratory flow was introduced as an external disturbance flow. Under these conditions, it was demonstrated that the respiratory state—i.e., the expiratory state (flow from the simulator to the pressure buffer) and the inspiratory state (flow from the pressure buffer to the simulator)—can be estimated from the differential pressure ΔP, the pump voltage V, and the PWM duty value d, with respect to the disturbance flow generated by the respiratory airflow simulator. It was also confirmed that such respiratory state estimation remains valid even when the duty value d of the pump is being actively modulated to control the internal pressure of the PAPR hood. Furthermore, based on the estimated respiratory states, a theoretical investigation was conducted on constant pressure control inside the PAPR and on the inverse pressure control aimed at supporting respiratory activity—namely, pressure control that assists breathing by depressurizing when expiratory motion is detected and pressurizing when inspiratory motion is detected. This study was conducted as part of a research and development project on public-oriented PAPR systems, which are being explored as alternatives to lockdown measures in response to airborne infectious diseases such as COVID-19. The present work specifically focused on improving the wearing comfort of the PAPR. Full article
(This article belongs to the Special Issue Sensors and Analog Front-End Circuits for Sensing Systems and High Sensitivity Measurements: 2nd Edition)
►▼ Show Figures

Figure 1

10 pages, 3365 KiB  
Article
Design of Small-Sized Spiral Slot PIFA Antenna Used Conformally in Laminated Body Tissues
by Rong Li, Jian Liu, Cuizhen Sun, Wang Yao, Ying Tian and Xiaojun Huang
Sensors 2025, 25(9), 2938; https://doi.org/10.3390/s25092938 - 7 May 2025
Viewed by 156
Abstract
This paper presents a novel Spiral Slot Planar Inverted-F Antenna (SSPIFA) specifically designed for telemedicine and healthcare applications, featuring compact size, biocompatible safety, and high integration suitability. By replacing the conventional top metal patch of a Planar Inverted-F Antenna (PIFA) with a slot [...] Read more.
This paper presents a novel Spiral Slot Planar Inverted-F Antenna (SSPIFA) specifically designed for telemedicine and healthcare applications, featuring compact size, biocompatible safety, and high integration suitability. By replacing the conventional top metal patch of a Planar Inverted-F Antenna (PIFA) with a slot spiral radiator whose geometry is precisely matched to the ground plane, the proposed antenna achieves a significant size reduction, making it ideal for encapsulation in miniaturized medical devices—a critical requirement for implantation scenarios. Tailored for the ISM 915 MHz band, the antenna is fabricated with a four-turn slot spiral etched on a 30 mm-diameter dielectric substrate, achieving an overall height of 22 mm and an electrically small profile of approximately 0.09λ × 0.06λ (λ: free-space wavelength at the center frequency). Simulation and measurement results demonstrate a −16 dB impedance matching (S11 parameter) at the target frequency, accompanied by a narrow fractional bandwidth of 1% and stable right-hand circular polarization (RHCP). When implanted in a layered biological tissue model (skin, fat, muscle), the antenna exhibits a near-omni directional radiation pattern in the azimuthal plane, with a peak gain of 2.94 dBi and consistent performance across the target band. These characteristics highlight the SSPIFA’s potential for reliable wireless communication in implantable medical systems, balancing miniaturization, radiation efficiency, and biocompatible design. Full article
(This article belongs to the Special Issue Metasurfaces for Enhanced Communication and Radar Detection)
►▼ Show Figures

Figure 1

17 pages, 8934 KiB  
Article
Real-Time Avalanche Hazard Monitoring System Based on Weather Sensors and a Laser Rangefinder
by Natalya Denissova, Olga Petrova, Erbolat Mashayev, Dmitry Spivak, Vitaly Zuyev and Gulzhan Daumova
Sensors 2025, 25(9), 2937; https://doi.org/10.3390/s25092937 - 7 May 2025
Viewed by 133
Abstract
Avalanche hazard prediction remains a crucial task for mountainous regions worldwide. This study presents the development and field testing of a prototype automated avalanche hazard monitoring system designed for the East Kazakhstan region. The system integrates a snow avalanche station (including temperature, humidity, [...] Read more.
Avalanche hazard prediction remains a crucial task for mountainous regions worldwide. This study presents the development and field testing of a prototype automated avalanche hazard monitoring system designed for the East Kazakhstan region. The system integrates a snow avalanche station (including temperature, humidity, and pressure sensors; a magnetoelectric wind sensor; a data logger; and devices for autonomous operation), a temperature snow measuring rod, an API (application programming interface) service for storing weather and climate parameters in a database, and a web interface. Powered by autonomous solar energy solutions, the system ensures continuous, high-resolution monitoring of key environmental parameters. Using initial test datasets, we analyzed the specific strengths and weaknesses of the developed monitoring system using the example of one avalanche site. Avalanche prediction was performed using regression analysis (logistic regression). The evaluation of the model showed a high forecasting accuracy, with recognition rates exceeding 98%. The obtained regression coefficients can be used to predict avalanches based on meteorological data collected using the proposed equipment. The developed solution holds significant promise for improving avalanche risk management practices and can be expanded for broader application in both national and international contexts. Full article
(This article belongs to the Section Optical Sensors)
►▼ Show Figures

Figure 1

13 pages, 5645 KiB  
Article
Morphology-Dependent Behavior of PVDF/ZnO Composites: Their Fabrication and Application in Pressure Sensors
by Binbin Zhang, Wenhui Zhang, Wei Luo, Zhijie Liang, Yan Hong, Jianhui Li, Guoyun Zhou and Wei He
Sensors 2025, 25(9), 2936; https://doi.org/10.3390/s25092936 - 7 May 2025
Viewed by 143
Abstract
This study investigated the impact of zinc oxide’s (ZnO’s) morphology on the piezoelectric performance of polyvinylidene fluoride (PVDF) composites for flexible sensors. Rod-like (NR) and sheet-like (NS) ZnO nanoparticles were synthesized via hydrothermal methods and incorporated into PVDF through direct ink writing (DIW). [...] Read more.
This study investigated the impact of zinc oxide’s (ZnO’s) morphology on the piezoelectric performance of polyvinylidene fluoride (PVDF) composites for flexible sensors. Rod-like (NR) and sheet-like (NS) ZnO nanoparticles were synthesized via hydrothermal methods and incorporated into PVDF through direct ink writing (DIW). The structural analyses confirmed the successful formation of wurtzite ZnO and enhanced β-phase content in the PVDF/ZnO composites. At a degree of 15 wt% loading, the ZnO-NS nanoparticles achieved the highest β-phase fraction (81.3%) in PVDF due to their high specific surface area, facilitating dipole alignment and strain-induced crystallization. The optimized PVDF/ZnO-NS-15 sensor demonstrated superior piezoelectric outputs (4.75 V, 140 mV/N sensitivity) under a 27 N force, outperforming its ZnO-NR counterparts (3.84 V, 100 mV/N). The cyclic tests revealed exceptional durability (<5% signal attenuation after 1000 impacts) and a rapid response (<100 ms). The application trials validated their real-time motion-monitoring capabilities, including finger joint flexion detection. This work highlights the morphology-dependent interfacial polarization as a critical factor for high-performance flexible sensors, offering a scalable DIW-based strategy for wearable electronics. Full article
(This article belongs to the Special Issue Functional Nanomaterials in Sensing)
►▼ Show Figures

Graphical abstract

21 pages, 6015 KiB  
Article
A Hybrid CBiGRUPE Model for Accurate Grinding Wheel Wear Prediction
by Sumei Si, Deqiang Mu and Hailiang Tang
Sensors 2025, 25(9), 2935; https://doi.org/10.3390/s25092935 - 6 May 2025
Viewed by 140
Abstract
In grinding machining, monitoring grinding wheel wear is essential for ensuring process quality wear and reducing production costs. This paper presents a hybrid CBiGRUPE model to predict grinding wheel wear, which integrates the advantages of convolutional neural networks (CNNs), bidirectional gated recurrent unit [...] Read more.
In grinding machining, monitoring grinding wheel wear is essential for ensuring process quality wear and reducing production costs. This paper presents a hybrid CBiGRUPE model to predict grinding wheel wear, which integrates the advantages of convolutional neural networks (CNNs), bidirectional gated recurrent unit (BiGRU), and the Performer encoder. Time-domain features are extracted from the spindle motor current signals of a surface grinding machine. The structure and hyperparameters of CBiGRUPE are optimized using Bayesian optimization. Experimental validation of the model demonstrates superior performance, with mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R2) values of 3.041, 3.927, and 0.920, respectively. Compared to models like CNN, BiGRU, and Transformer, the CBiGRUPE model offers more accurate and stable wear predictions. This paper also discusses the advantages and limitations of various models for estimating grinding wheel wear, emphasizing the effectiveness of the proposed approach. This study establishes a foundation for compensating wheel wear and accurately determining the optimal dressing time. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

15 pages, 7160 KiB  
Article
Dual-Band Dual-Beam Shared-Aperture Reflector Antenna Design with FSS Subreflector
by Qunbiao Wang, Peng Li, Guodong Tan, Yiqun Zhang, Yuanxin Yan, Wanye Xu and Paolo Rocca
Sensors 2025, 25(9), 2934; https://doi.org/10.3390/s25092934 - 6 May 2025
Viewed by 164
Abstract
In this study, a dual-band dual-beam shared-aperture reflector antenna based on a Cassegrain configuration is designed using a frequency-selective surface (FSS) subreflector. The antenna generates two shaped beams that operate at different frequencies and can spatially overlap. One beam contour can be independently [...] Read more.
In this study, a dual-band dual-beam shared-aperture reflector antenna based on a Cassegrain configuration is designed using a frequency-selective surface (FSS) subreflector. The antenna generates two shaped beams that operate at different frequencies and can spatially overlap. One beam contour can be independently optimized by properly designing the shape of the main reflector. The contour of the second beam is defined by optimizing the unit cell and geometry of the FSS-based subreflector once the shape of the main reflector is set. The reflector antenna design is cast as the optimization of a suitably defined cost function aimed at yielding the desired directivity performance in the regions of coverage. In order to validate the proposed solution, a set of numerical experiments was conducted using most of China and Shaanxi province as benchmark examples. Full article
(This article belongs to the Special Issue Sensors in 2025)
►▼ Show Figures

Figure 1

12 pages, 16116 KiB  
Article
All-Fiber LITES Sensor Based on Hollow-Core Anti-Resonant Fiber and Self-Designed Low-Frequency Quartz Tuning Fork
by Xiaorong Sun, Weipeng Chen, Ying He, Haiyue Sun, Shunda Qiao and Yufei Ma
Sensors 2025, 25(9), 2933; https://doi.org/10.3390/s25092933 - 6 May 2025
Viewed by 175
Abstract
In this paper, an all-fiber light-induced thermoelastic spectroscopy (LITES) sensor based on hollow-core anti-resonant fiber (HC-ARF) and self-designed low-frequency quartz tuning fork (QTF) is reported for the first time. By utilizing HC-ARF as both the transmission medium and gas chamber, the laser tail [...] Read more.
In this paper, an all-fiber light-induced thermoelastic spectroscopy (LITES) sensor based on hollow-core anti-resonant fiber (HC-ARF) and self-designed low-frequency quartz tuning fork (QTF) is reported for the first time. By utilizing HC-ARF as both the transmission medium and gas chamber, the laser tail fiber was spatially coupled with the HC-ARF, and the end of the HC-ARF was directly guided onto the QTF surface, resulting in an all-fiber structure. This design eliminated the need for lens combinations, thereby enhancing system stability and reducing cost and size. Additionally, a self-designed rectangular-tip QTF with a low resonant frequency of 8.69 kHz was employed to improve the sensor’s detection performance. Acetylene (C2H2), with an absorption line at 6534.37 cm−1 (1.53 μm), was chosen as the target gas. Experimental results clearly demonstrated that the detection performance of the rectangular-tip QTF system was 2.9-fold higher than that of a standard commercial QTF system. Moreover, it exhibited an outstanding linear response to varying C2H2 concentrations, indicating its high sensitivity and reliability in detecting C2H2. The Allan deviation analysis was used to assess the system’s stability, and the results indicated that the system exhibits excellent long-term stability. Full article
(This article belongs to the Special Issue Applications of Optical Sensing and Laser Spectroscopy in Gas Detection)
►▼ Show Figures

Figure 1

16 pages, 4262 KiB  
Article
Multimodal MRI Image Fusion for Early Automatic Staging of Endometrial Cancer
by Ziyu Zheng, Ye Liu, Longxiang Feng, Peizhong Liu, Haisheng Song, Lin Wang and Fang Huang
Sensors 2025, 25(9), 2932; https://doi.org/10.3390/s25092932 - 6 May 2025
Viewed by 238
Abstract
This magnetic resonance imaging multimodal fusion study aims to automate the staging of endometrial cancer using deep learning and to compare the diagnostic performance of deep learning with that of radiologists in the staging of endometrial cancer. This study retrospectively investigated 122 patients [...] Read more.
This magnetic resonance imaging multimodal fusion study aims to automate the staging of endometrial cancer using deep learning and to compare the diagnostic performance of deep learning with that of radiologists in the staging of endometrial cancer. This study retrospectively investigated 122 patients with pathologically confirmed early EC from January 1, 2025 to December 31, 2021. Of these patients, 68 were in the International Federation of Gynecology and Obstetrics (FIGO) stage IA, and 54 were in FIGO stage IB. Based on the Swin transformer model and its proprietary SW-MSA (shift window multiple self-coherence) module, magnetic resonance imaging (MRI) images in each of the three planes (sagittal, coronal, and transverse) are cropped, enhanced, and classified, and fusion experiments in the three planes are performed simultaneously. Selecting one plane for the experiment, the accuracy of IA and IB classification was 0.988 in the sagittal, 0.96 in the coronal, and 0.94 in the transverse position, and classification accuracy after the fusion of three planes reached 1. Finally, the automatic classification method based on the Swin transformer has an accuracy of 1, a recall of 1, and a specificity of 1 for early EC classification. In this study, the multimodal fusion approach accurately classified early EC. It was comparable to what a radiologist would perform and simpler and more precise than previous methods that required segmenting followed by staging. Full article
(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments, 2nd Edition)
►▼ Show Figures

Figure 1

18 pages, 3197 KiB  
Article
An Open-Source Wearable System for Real-Time Human Biomechanical Analysis
by Zachary Hoegberg, Seth Donahue and Matthew J. Major
Sensors 2025, 25(9), 2931; https://doi.org/10.3390/s25092931 - 6 May 2025
Viewed by 230
Abstract
The advancement of inertial measurement unit (IMU) technology has opened new opportunities for motion analysis, yet its widespread adoption in clinical practice remains constrained by the high costs of proprietary systems, lengthy setup procedures, and the need for specialized expertise. To address these [...] Read more.
The advancement of inertial measurement unit (IMU) technology has opened new opportunities for motion analysis, yet its widespread adoption in clinical practice remains constrained by the high costs of proprietary systems, lengthy setup procedures, and the need for specialized expertise. To address these challenges, we present a multi-IMU system designed with streamlined calibration, efficient data processing, and a focus on accessibility for patient-facing applications. Although initially developed for human gait analysis, the modular design of this system enables adaptability across diverse motion tracking scenarios. This work outlines the system’s technical framework, including protocols for data acquisition, derivation of gait variables, and considerations for user-friendly software deployment. We further illustrate its utility by measuring lower-limb gait kinematics in near-real time and providing stride-to-stride biofeedback using a single sensor. These initial results underscore the potential of this system for both laboratory-based gait assessment and rehabilitation interventions in clinical environments and future work will assess validation against traditional optical motion capture methods. Full article
(This article belongs to the Special Issue Innovative Applications of Wearable Sensors in Musculoskeletal Biomechanics and Rehabilitation)
►▼ Show Figures

Figure 1

19 pages, 30474 KiB  
Article
Multi-Head Attention-Based Framework with Residual Network for Human Action Recognition
by Basheer Al-Tawil, Magnus Jung, Thorsten Hempel and Ayoub Al-Hamadi
Sensors 2025, 25(9), 2930; https://doi.org/10.3390/s25092930 - 6 May 2025
Viewed by 226
Abstract
Human action recognition (HAR) is essential for understanding and classifying human movements. It is widely used in real-life applications such as human–computer interaction and assistive robotics. However, recognizing patterns across different temporal scales remains challenging. Traditional methods struggle with complex timing patterns, intra-class [...] Read more.
Human action recognition (HAR) is essential for understanding and classifying human movements. It is widely used in real-life applications such as human–computer interaction and assistive robotics. However, recognizing patterns across different temporal scales remains challenging. Traditional methods struggle with complex timing patterns, intra-class variability, and inter-class similarities, leading to misclassifications. In this paper, we propose a deep learning framework for efficient and robust HAR. It integrates residual networks (ResNet-18) for spatial feature extraction and Bi-LSTM for temporal feature extraction. A multi-head attention mechanism enhances the prioritization of crucial motion details. Additionally, we introduce a motion-based frame selection strategy utilizing optical flow to reduce redundancy and enhance efficiency. This ensures accurate, real-time recognition of both simple and complex actions. We evaluate the framework on the UCF-101 dataset, achieving a 96.60% accuracy, demonstrating competitive performance against state-of-the-art approaches. Moreover, the framework operates at 222 frames per second (FPS), achieving an optimal balance between recognition performance and computational efficiency. The proposed framework was also deployed and tested on a mobile service robot, TIAGo, validating its real-time applicability in real-world scenarios. It effectively models human actions while minimizing frame dependency, making it well-suited for real-time applications. Full article
(This article belongs to the Special Issue Deep Learning Applications for Pose Estimation and Human Action Recognition—2nd Edition)
►▼ Show Figures

Figure 1

22 pages, 6945 KiB  
Article
Parameter Calculation and Rotor Structure Optimization Design of Solid Rotor Induction Motors
by Hao Xu, Jinghong Zhao and Sinian Yan
Sensors 2025, 25(9), 2929; https://doi.org/10.3390/s25092929 - 6 May 2025
Viewed by 217
Abstract
Solid rotor induction motors have a solid body rotor, which leads to low efficiency and power factor, and currently, the rotor is mostly optimized by slotted and squirrel cage structures. A generalized multilayer analytical model for different rotor structures is established, which can [...] Read more.
Solid rotor induction motors have a solid body rotor, which leads to low efficiency and power factor, and currently, the rotor is mostly optimized by slotted and squirrel cage structures. A generalized multilayer analytical model for different rotor structures is established, which can consider the effects of rotor eddy currents and saturation, based on which a generalized equivalent circuit model is established. The effects of number of slots, depth of slots, width of slots, squirrel cage material and end ring thickness on rotor impedance, torque and rotor losses are analyzed. On this basis, with efficiency, power factor, starting torque and starting current as the optimization objectives, and the number of slots, slot depth, slot width, squirrel cage material and end ring thickness as the optimization variables, the optimization schemes of slotted rotor and squirrel cage rotor are obtained by using the three-dimensional finite element method. The theoretical analysis is verified by finite element simulation and prototype experiment, and the results show that the electromagnetic parameters of solid rotor induction motors with different rotor structures can be accurately calculated using the universal magnetic field analytical model and the universal equivalent circuit model with an error within 5.8%. Slotted and squirrel cage rotors can effectively improve the motor power factor and efficiency, but this will lead to a decrease in starting performance. For the optimization function established in this paper, compared with the smooth rotor, the performance of the squirrel cage rotor is improved by 6.08%, which verifies the accuracy and validity of this paper and the optimization design scheme. Full article
(This article belongs to the Special Issue Recent Trends in AI-Based Intelligent Sensing Systems and IoTs)
►▼ Show Figures

Figure 1

13 pages, 2679 KiB  
Article
Terahertz Frequency-Modulated Continuous-Wave Inspection of an Ancient Enamel Plate
by Frédéric Fauquet, Francesca Galluzzi, Rémy Chapoulie, Aurélie Mounier, Ayed Ben Amara and Patrick Mounaix
Sensors 2025, 25(9), 2928; https://doi.org/10.3390/s25092928 - 6 May 2025
Viewed by 217
Abstract
This study investigates the application of terahertz frequency-modulated continuous-wave (FMCW) imaging for the non-destructive inspection of a historical enamel plate, using both reflection and transmission modes. A 300 GHz FMCW radar system was employed to capture high-resolution images of the plate’s internal and [...] Read more.
This study investigates the application of terahertz frequency-modulated continuous-wave (FMCW) imaging for the non-destructive inspection of a historical enamel plate, using both reflection and transmission modes. A 300 GHz FMCW radar system was employed to capture high-resolution images of the plate’s internal and surface structures. Through optimized data acquisition and processing, the system successfully revealed subsurface features such as fractures, as well as surface-level textural variations linked to the decorative glazes. Although pigment differentiation remains a challenge, contrast variations observed in THz images suggest correlations with material composition. The results highlight the potential of FMCW terahertz imaging as a compact, rapid, and non-contact diagnostic tool for cultural heritage analysis. Its practicality and adaptability make it particularly suitable for in situ inspections in museums or restoration contexts. Full article
(This article belongs to the Special Issue Recent Advances in THz Sensing and Imaging)
►▼ Show Figures

Figure 1

19 pages, 8785 KiB  
Article
Design of a 5G MIMO Mobile Intelligent Terminal Antenna with Metasurface Loading
by He Xia, Heming Fan, Zhulin Liu, Hongxiang Miao and Zhiwei Song
Sensors 2025, 25(9), 2927; https://doi.org/10.3390/s25092927 - 6 May 2025
Viewed by 234
Abstract
To achieve multi-band coverage within limited space, reduce antenna types, and enhance communication capabilities, an eight-unit dual-band 5G MIMO antenna array is proposed based on a monopole structure. The antenna operates in two frequency bands (3.23–4.14 GHz and 4.31–5.3 GHz), covering the n78 [...] Read more.
To achieve multi-band coverage within limited space, reduce antenna types, and enhance communication capabilities, an eight-unit dual-band 5G MIMO antenna array is proposed based on a monopole structure. The antenna operates in two frequency bands (3.23–4.14 GHz and 4.31–5.3 GHz), covering the n78 and n79 bands for 5G applications. The dual-band and miniaturized design of the antenna elements is achieved through the slotting and branch-loading techniques. The orthogonal placement of corner antenna elements is implemented to reduce coupling and optimize spatial utilization, achieving isolation of over 16 dB between elements. The introduction of a metasurface structure further improved isolation by 2 dB and increased the peak gain of the antenna array to 11.95 dBi. A prototype is fabricated and tested, demonstrating the following performance metrics: isolation exceeding 18 dB, gain ranging from 6 to 12 dBi, envelope correlation coefficient below 0.05, channel capacity greater than 41 bps/Hz, diversity gain of approximately 10 dB, total active reflection coefficient below −24 dB, and radiation efficiency exceeding 72%. These results confirm the superior performance of the proposed antenna design. Full article
(This article belongs to the Special Issue Advanced Massive MIMO Antenna Arrays, Metasurfaces and Reconfigurable Intelligent Surfaces for Sensing, Localization, and Wireless Communications: 2nd Edition)
►▼ Show Figures

Figure 1

15 pages, 3478 KiB  
Article
Validation of an Open-Source Smartwatch for Continuous Monitoring of Physical Activity and Heart Rate in Adults
by Nicholas Ravanelli, KarLee Lefebvre, Amy Brough, Simon Paquette and Wei Lin
Sensors 2025, 25(9), 2926; https://doi.org/10.3390/s25092926 - 6 May 2025
Viewed by 299
Abstract
Consumer-grade wrist-based wearable devices have grown in popularity among researchers to continuously collect metrics such as physical activity and heart rate. However, manufacturers rarely disclose the preprocessing sensor data algorithms, and user-generated data are typically shared leading to data governance issues. Open-source technology [...] Read more.
Consumer-grade wrist-based wearable devices have grown in popularity among researchers to continuously collect metrics such as physical activity and heart rate. However, manufacturers rarely disclose the preprocessing sensor data algorithms, and user-generated data are typically shared leading to data governance issues. Open-source technology may address these limitations. This study evaluates the validity of the Bangle.js2 for step counting and heart rate during lab-based validation and agreement with other wearable devices (steps: Fitbit Charge 5; heart rate: Polar H10) in free-living conditions. A custom open-source application was developed to capture the sensor data from the Bangle.js2. Participants (n = 47; 25 males; 27 ± 11 years) were asked to complete a lab-based treadmill validation (3 min stages at 2, 3, 4, and 5 mph) and stair climbing procedure followed by a 24 h free-living period. The Bangle.js2 demonstrated systematic undercounting of steps at slower walking speeds with acceptable error achieved at 5 km/h. During free-living conditions, the Bangle.js2 demonstrated strong agreement with the Fitbit Charge 5 for per-minute step counting (CCC = 0.90) and total steps over 24 h (CCC = 0.96). Additionally, the Bangle.js2 demonstrated strong agreement with the Polar H10 for minute-averaged heart rate (CCC = 0.78). In conclusion, the Bangle.js2 is a valid open-source hardware and software solution for researchers interested in step counting and heart rate monitoring in free-living conditions. Full article
(This article belongs to the Special Issue Wearable Sensing Technology for Physiological and Behavioral Human Monitoring—2nd Edition)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-322
Previous Issue
Next Issue
Back to TopTop