Sensors

14 pages, 693 KB

Open AccessArticle

Benchmarking Time-Series Artificial Intelligence Architectures for Wearable Sensor-Based Fall Prediction: A Synthetic Data Simulation Framework

by Edward R. Sykes, Mohammad Maghsoudimehrabani and Abdulrahman Al-Shanoon

Sensors 2026, 26(11), 3326; https://doi.org/10.3390/s26113326 (registering DOI) - 24 May 2026

Abstract

Falls among older adults are a major cause of injury and loss of independence, yet most existing systems detect falls only after onset or provide very limited warning time. This study presents a synthetic benchmarking framework for early fall-risk prediction using multimodal wearable-inspired [...] Read more.

Falls among older adults are a major cause of injury and loss of independence, yet most existing systems detect falls only after onset or provide very limited warning time. This study presents a synthetic benchmarking framework for early fall-risk prediction using multimodal wearable-inspired time-series data and compares classical and temporal machine learning architectures under a realistic evaluation protocol. A synthetic dataset of 1000 sequences was generated to emulate normal activity, slip events, and pre-fall instability using biomechanical, physiological, and contextual variables. Eight baseline models and two augmented temporal variants were trained and evaluated using subject-wise splits to reduce leakage. Performance differed substantially by model family and evaluation protocol. Classical baselines achieved the strongest overall macro-F1 scores, whereas temporal models showed more modest discrimination. Under a fixed alerting rule, operational early-warning behavior varied considerably: some models failed to trigger alerts, while others achieved higher pre-fall trigger rates at the cost of increased false alarms. These findings show that apparent performance depends strongly on partitioning strategy, calibration, and alert design. The proposed framework provides a reproducible basis for benchmarking early-warning fall-risk models and supports future validation using real-world cohorts and deployment-oriented calibration strategies. Full article

(This article belongs to the Special Issue Sensor-Based Wearable Devices: Health, Activity and Monitoring)

► Show Figures

Figure 1

23 pages, 2472 KB

Open AccessArticle

Stability-Controlled Continual Federated Learning for Energy-Harvesting AIoT Systems

by Junsoo Park, Ikjune Yoon and Dong Kun Noh

Sensors 2026, 26(11), 3325; https://doi.org/10.3390/s26113325 (registering DOI) - 23 May 2026

Abstract

Energy-harvesting (EH) AIoT systems enable long-term autonomous operation but suffer from time-varying energy availability, which makes stable learning difficult. In such environments, federated learning (FL) is prone to energy depletion (blackout), while continual learning is required to handle evolving data distributions, leading to [...] Read more.

Energy-harvesting (EH) AIoT systems enable long-term autonomous operation but suffer from time-varying energy availability, which makes stable learning difficult. In such environments, federated learning (FL) is prone to energy depletion (blackout), while continual learning is required to handle evolving data distributions, leading to a trade-off between energy stability and catastrophic forgetting. In this paper, we propose a stability-controlled continual federated learning framework that jointly regulates local training intensity and rehearsal usage based on the residual energy state. The proposed method is derived from a Lyapunov drift-plus-penalty formulation and implemented as a lightweight mode-based control policy. Simulation results using real solar energy traces show that the proposed method significantly reduces blackout while improving accuracy and mitigating forgetting compared to existing approaches. These results demonstrate the effectiveness of energy-aware joint control for stable continual federated learning in EH-AIoT systems. Full article

(This article belongs to the Special Issue New Trends in Artificial Intelligence of Things (AIoT))

34 pages, 3599 KB

Open AccessSystematic Review

CD4⁺ T Lymphocytes—From Their Multiple Immunological Roles to New Sensing Strategies Using Marine Magnetotactic Bacteria—A Systematic Review

by Natalia Lorela Paul, Catalin Ovidiu Popa and Rodica Elena Ionescu

Sensors 2026, 26(11), 3324; https://doi.org/10.3390/s26113324 (registering DOI) - 23 May 2026

Abstract

CD4⁺ T lymphocytes are important regulators of immune homeostasis, and their dysregulation is closely linked to a wide range of diseases. For this reason, their reliable detection remains a major challenge. Despite the fact that current methods are analytically robust, they rely [...] Read more.

CD4⁺ T lymphocytes are important regulators of immune homeostasis, and their dysregulation is closely linked to a wide range of diseases. For this reason, their reliable detection remains a major challenge. Despite the fact that current methods are analytically robust, they rely mainly on laboratory infrastructure, limiting their flexibility and wider applicability. The present review analyzes established and emerging approaches for CD4⁺ T-cell detection, with a focus on their practical limitations regarding portability, flexibility and applicability. To the best of our knowledge, for the first time, the authors are examining the possibility of exploring marine magnetotactic bacteria (MTBs) as active biological elements in immune cell detection due to their intrinsic magnetic properties, biological organization, and surface biofunctionalization potential. Rather than offering an immediate technological solution, the use of MTBs serves as a challenging biological framework that could provide more adaptable and sensitive detection strategies. At the same time, the limitations of this concept are acknowledged, emphasizing the need for further experimental validation, considering that this strategy, although promising, remains an exploratory concept. Full article

(This article belongs to the Section Biosensors)

15 pages, 544 KB

Open AccessArticle

Air Target ISAR Recognition Based on Data Augmentation and Transfer Learning

by Moqian Wang, Zuzhen Huang, Jinjian Cai, Tao Wu and Youquan Lin

Sensors 2026, 26(11), 3323; https://doi.org/10.3390/s26113323 (registering DOI) - 23 May 2026

Abstract

Aiming at the problems of extremely scarce measured samples and significant domain shift between simulated and measured data in automatic target recognition (ATR) of air targets for spaceborne radar, this paper proposes an inverse synthetic aperture radar (ISAR) image recognition method for air [...] Read more.

Aiming at the problems of extremely scarce measured samples and significant domain shift between simulated and measured data in automatic target recognition (ATR) of air targets for spaceborne radar, this paper proposes an inverse synthetic aperture radar (ISAR) image recognition method for air targets combining physics-driven data augmentation guided by detection prior information with domain adversarial transfer learning. First, the mapping relationship between scattering point projection and ISAR images is established by using the target 3D point cloud and radar observation geometric priors, and a 2D sinc kernel function is introduced for energy distribution rendering. Then, under the unsupervised transfer learning paradigm, aiming at the distribution inconsistency between augmented data (source domain) and unlabeled simulated data (target domain), this paper designs a cross-domain recognition task experiment including six types of typical aircraft targets, and compares the cross-domain recognition performance of three transfer learning methods (model fine-tuning, deep domain confusion (DDC) and domain-adversarial neural networks (DANN)) on the target domain. Meanwhile, t-distributed stochastic neighbor embedding (t-SNE) visualization is used to analyze the feature distribution alignment ability of the models. Simulation experiments show that the DANN model with a dynamic inversion coefficient introduced in the gradient reversal layer (GRL) achieves a recognition accuracy of 99.5% on the unlabeled target domain, which is significantly superior to the model fine-tuning and DDC methods. Moreover, it makes the feature distributions of source and target domain samples highly overlapping, and maintains a strong inter-class discriminability while eliminating the domain shift. The proposed scheme provides a physically interpretable and robust technical path for few-shot radar target image recognition. Full article

(This article belongs to the Section Radar Sensors)

24 pages, 1809 KB

Open AccessFeature PaperArticle

Cloud-to-Edge Deployment of Optimized nnU-Net for Ischemic Stroke Lesion Segmentation on Resource-Constrained Embedded Devices

by Daniel Alcaraz Ortiz, Juan Francisco Zapata Pérez and Juan Martinez-Alajarin

Sensors 2026, 26(11), 3322; https://doi.org/10.3390/s26113322 (registering DOI) - 23 May 2026

Abstract

Ischemic stroke remains a leading cause of global mortality and long-term neurological disability, where the “Time is Brain” paradigm dictates that rapid and accurate lesion assessment is fundamental for effective clinical intervention. While the nnU-Net v2 framework has established a new state of [...] Read more.

Ischemic stroke remains a leading cause of global mortality and long-term neurological disability, where the “Time is Brain” paradigm dictates that rapid and accurate lesion assessment is fundamental for effective clinical intervention. While the nnU-Net v2 framework has established a new state of the art in medical image segmentation, its high computational demands and reliance on data-center-grade GPUs hinder its translation into real-time, point-of-care clinical workflows. This study presents a technical feasibility analysis of a Cloud-to-Edge optimization pipeline designed to transfer a 3D nnU-Net v2 model from a high-performance cloud environment to a resource-constrained embedded device. Experimental results showed that edge deployment was associated with a reduction in overlap-based segmentation metrics compared with the cloud reference, with Dice decreasing from approximately 0.78 to 0.67. However, TensorRT FP32 and FP16 inference produced nearly identical mean segmentation metrics, suggesting that reduced-precision inference did not introduce additional measurable degradation under the evaluated conditions. The optimized FP16 configuration achieved a processing time of 10.2 s per 3D volume, representing a 33% reduction compared with embedded FP32 inference, while operating within a low-power envelope of approximately 10–13 W. These findings support the preliminary technical feasibility of executing advanced 3D volumetric segmentation models on low-power edge hardware. Nevertheless, the evaluation was limited to an internal 25-case test subset and did not include external validation, prospective clinical assessment, or reader studies. Therefore, the proposed system should be interpreted as a preliminary deployment framework rather than a clinically validated tool for autonomous stroke imaging. Full article

(This article belongs to the Special Issue Deep Learning and Artificial Intelligence Methods for Diagnostics in Medical Imaging and Sensing)

► Show Figures

Figure 1

22 pages, 3661 KB

Open AccessArticle

Industrial Weld Defect Detection Based on Monocular Depth Estimation and Dual-Attention Point Cloud Network

by Nannan Zhao and Shijie Chen

Sensors 2026, 26(11), 3321; https://doi.org/10.3390/s26113321 (registering DOI) - 23 May 2026

Abstract

In industrial quality control, the precise identification of severe structural weld defects is paramount. Traditional 2D image-based detection methods are susceptible to illumination and texture interference, while high-precision 3D laser scanning solutions are costly and impractical for large-scale deployment. To achieve reliable geometric [...] Read more.

In industrial quality control, the precise identification of severe structural weld defects is paramount. Traditional 2D image-based detection methods are susceptible to illumination and texture interference, while high-precision 3D laser scanning solutions are costly and impractical for large-scale deployment. To achieve reliable geometric defect detection at low cost, this paper proposes a detection framework based on monocular depth estimation and a dual-attention point cloud network. First, YOLOv8 is employed for rapid region of interest extraction, and an advanced monocular depth estimation model generates 3D pseudo-point clouds containing geometric information. Secondly, addressing the challenge of distinct spatial orientation features in missed weld defects that are prone to confusion, this paper introduces a dual-attention-enhanced point cloud classification network named DA-PointNet++. This model embeds dual-attention modules within the PointNet++ backbone network, enhancing key feature representation in both the channel and spatial dimensions. Experimental results demonstrate that this approach achieves an accuracy of 93.67% and a recall rate of 90.51% in a unified binary classification task for general weld defect detection, effectively identifying both normal welds and complex missed weld defects. Compared to PointConv, Dynamic Graph Convolutional Neural Network (DGCNN), and mainstream Point Cloud Transformer, this method significantly reduces false negative rates while maintaining low computational costs, offering a cost-effective solution for industrial automation. Full article

(This article belongs to the Section Industrial Sensors)

► Show Figures

Figure 1

21 pages, 4832 KB

Open AccessArticle

YOLOv9-Based Detection of Diseases in Poplar Trees Using Histogram Equalization and Computer Vision

by Fazliddin Makhmudov, Kudratjon Zohirov, Jura Kuvandikov, Zavqiddin Temirov, Akmalbek Abdusalomov Bobomirzayevich, Mukhriddin Mukhiddinov, Khodisakhon Muraeva, Jasur Sevinov and Furkat Bolikulov

Sensors 2026, 26(11), 3320; https://doi.org/10.3390/s26113320 (registering DOI) - 23 May 2026

Abstract

Poplar (Populus) trees are indispensable to various industries and environmental sustainability efforts. They are widely utilized for paper production, timber, and windbreaks, while also playing a significant role in carbon sequestration. Given their economic and ecological importance, the effective management of diseases is [...] Read more.

Poplar (Populus) trees are indispensable to various industries and environmental sustainability efforts. They are widely utilized for paper production, timber, and windbreaks, while also playing a significant role in carbon sequestration. Given their economic and ecological importance, the effective management of diseases is crucial. Convolutional Neural Networks (CNNs), renowned for their ability to process visual data, are pivotal in accurately detecting and classifying plant diseases. This study presents a domain-specific dataset of manually collected images of diseased poplar leaves from Uzbekistan and South Korea, ensuring geographic diversity and broader applicability. The dataset includes four disease classes, i.e., “Parsha (Scab),” “Brown spotting,” “White-Gray spotting,” and “Rust,” which represent common afflictions in these regions. To advance research efforts, this dataset will be made publicly accessible, providing a valuable resource for the scientific community. Leveraging the cutting-edge YOLOv9c model, a state-of-the-art CNN architecture, we applied the Histogram Equalization technique as a preprocessing step to enhance the image quality to increase the accuracy of disease detection. This method not only improves the diagnostic performance of the model but also provides a scalable solution for monitoring and managing poplar diseases. By ensuring the health of poplar trees, this approach supports the sustainability of these critical resources. To our knowledge, this is the first publicly available dataset specifically focused on diseased poplar leaves, making it a significant contribution to global research efforts. It offers an invaluable resource for researchers and practitioners, enabling further advancements in early disease detection and sustainable forestry management. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

16 pages, 729 KB

Open AccessArticle

Preoperative Decline and Postoperative Recovery of Wearable-Derived Physical Activity over a Four-Year Perioperative Period in Total Knee and Hip Arthroplasty

by Yuezhou Zhang, Amos Folarin, Rongrong Zhong, Hyunju Kim, Callum Stewart, Shaoxiong Sun and Richard J. B. Dobson

Sensors 2026, 26(11), 3319; https://doi.org/10.3390/s26113319 (registering DOI) - 23 May 2026

Abstract

We characterized long-term, objectively measured physical activity trajectories surrounding total knee arthroplasty (TKA) and total hip arthroplasty (THA) and examined factors associated with wearable-derived physical activity recovery. In this observational study within the All of Us Research Program, linked electronic health records and [...] Read more.

We characterized long-term, objectively measured physical activity trajectories surrounding total knee arthroplasty (TKA) and total hip arthroplasty (THA) and examined factors associated with wearable-derived physical activity recovery. In this observational study within the All of Us Research Program, linked electronic health records and Fitbit step count data spanning the two years before and the two years after surgery were analyzed using piecewise linear mixed-effects models to characterize preoperative and postoperative trajectories. Recovery of physical activity was defined relative to two preoperative baselines—activity measured immediately before surgery and a more remote baseline approximating longer-term habitual activity—and associated factors were examined using Cox proportional hazards models. Among 238 participants (147 TKA, 91 THA; mean age 64.9 [SD 8.3] years), both procedures showed progressive preoperative decline, with accelerated decline beginning earlier in TKA than in THA. Postoperative recovery followed a staged pattern, with rapid early improvement, slower intermediate gains, and later stabilization. Recovery to the immediate preoperative baseline occurred earlier than recovery to the remote baseline. Higher activity during the 4 weeks before surgery was associated with a greater likelihood of recovery to the remote baseline. These findings support long-term wearable monitoring as a complementary measure of physical activity recovery after arthroplasty. Full article

(This article belongs to the Special Issue Sensing Technologies for Mobile Health Monitoring)

12 pages, 1042 KB

Open AccessArticle

Measurement of Dermal Ammonia Emission Using a Passive Flux Sampler and Its Association with Autonomic Nervous System Activity in Medical Workers: A Preliminary Study

by Satomi Asai, Shiro Ikeda, Masaru Shiraiwa, Noboru Takanashi, Kazuo Umezawa, Kentaro Wakamatsu and Yoshika Sekine

Sensors 2026, 26(11), 3318; https://doi.org/10.3390/s26113318 (registering DOI) - 23 May 2026

Abstract

Medical workers are frequently exposed to high-stress environments, highlighting the need for non-invasive stress monitoring strategies based on autonomic nervous system activity. Ammonia emitted from the human skin surface has been reported to increase under physical and psychological stress; however, its relationship with [...] Read more.

Medical workers are frequently exposed to high-stress environments, highlighting the need for non-invasive stress monitoring strategies based on autonomic nervous system activity. Ammonia emitted from the human skin surface has been reported to increase under physical and psychological stress; however, its relationship with autonomic nervous system (ANS) dynamics remains unclear. In this study, we performed simultaneous, time-resolved measurements of dermal ammonia emission and heart rate variability (HRV) in 11 medical workers during 3 h of routine work. Dermal ammonia emission flux was continuously monitored using a passive flux sampler (PFS) coupled with ion chromatography, while autonomic nervous system activity was assessed by Holter electrocardiography. The temporal profiles of ammonia emission were analyzed in relation to HRV indices, including high frequency (HF) and the low-frequency-to-high-frequency ratio (LF/HF). Dermal ammonia emission increased under conditions characterized by lower HF and/or higher LF/HF, whereas elevated HF was associated with reduced ammonia emission (r = −0.47, p < 0.001). Furthermore, temporal fluctuations in ammonia emission were associated with sympathetic–parasympathetic switching. These findings suggest that dermal ammonia emission may be associated with HRV-related physiological responses under real-world working conditions and may have potential as a non-invasive indicator for stress-related physiological monitoring. Full article

(This article belongs to the Special Issue Advances in Gas Sensing Technologies for Disease Diagnosis Application)

► Show Figures

Figure 1

26 pages, 7009 KB

Open AccessArticle

Design, Implementation, and Verification of High-Accuracy Trapezoidal Dual-Axis Sun Sensors for LEO Satellite Attitude Determination

by Mang Ou-Yang, Ching-I Tai, Guan-Yu Huang, Tse-Yu Cheng, Chang-Hsun Liu, Yu-Siou Liu, Jin-Chern Chiou, Chen-Yu Chan, Tung-Yun Hsieh, Chen-Tsung Lin, Ying-Wen Jan, Chih-Hsun Lin and Yung-Jhe Yan

Sensors 2026, 26(11), 3317; https://doi.org/10.3390/s26113317 (registering DOI) - 23 May 2026

Abstract

This paper presents a dual-axis sun sensor employing a cross-slit aperture in conjunction with a four-quadrant trapezoidal photodiode layout. The cross-slit configuration enhances angular sensitivity and resolution, while the trapezoidal photodiode geometry preserves a high signal-to-noise ratio at both near-normal incidence and large [...] Read more.

This paper presents a dual-axis sun sensor employing a cross-slit aperture in conjunction with a four-quadrant trapezoidal photodiode layout. The cross-slit configuration enhances angular sensitivity and resolution, while the trapezoidal photodiode geometry preserves a high signal-to-noise ratio at both near-normal incidence and large Sun angles, maintaining reliable directional discriminability around normal incidence. Compared with conventional quad-triangle photodiode layouts, the proposed trapezoidal geometry avoids the rapid collapse of the illuminated area near the triangular apex at large incidence angles, thereby preserving signal margin near the field-of-view boundary. System-level optical verification demonstrates that, after calibration, the proposed sensor achieves an angular accuracy of

\pm 0.3

^{\circ}

(3

σ

). To mitigate performance variations induced by temperature drift, an embedded shielded dummy photodiode is incorporated to provide a dark-current reference for compensation. Unlike compensation approaches that mainly rely on pre-characterization or offline calibration, the embedded shielded dummy photodiode provides an in situ, real-time dark-current reference for compensating for temperature-induced signal drift in the actual operating environment. Experimental results under dark conditions indicate that the embedded dummy photodiode served as a dark-current reference for compensating the temperature-dependent dark-current variation in the active photodiodes, reducing the peak-to-peak dark-signal variation by 96% over a temperature range from 20

^{\circ} C

–120

^{\circ} C

. Furthermore, a pyramid-type sun-sensor architecture is proposed by integrating the dual-axis fine sun sensor with four wide-field coarse sun sensors. This system-level configuration extends the effective Sun field of view from the conventional 120

^{\circ}

– 180

^{\circ}

range to approximately 280

^{\circ}

, enabling near-hemispherical Sun-angle observability for enhanced attitude determination robustness. Full article

(This article belongs to the Special Issue Advanced Sensing, Control, Autonomous and Fault Tolerant Technologies for Satellite Applications)

20 pages, 1394 KB

Open AccessArticle

Unfolded RPCA Network for Mitigating Inter-Transmitter Code Interference in MIMO PMCW Systems

by Yonghee Lee, Jong-Ho Lee and Seongwook Lee

Sensors 2026, 26(11), 3316; https://doi.org/10.3390/s26113316 (registering DOI) - 23 May 2026

Abstract

Phase-modulated continuous wave (PMCW) has emerged as a promising waveform candidate for next-generation integrated sensing and communication systems due to its favorable sensing performance and multiplexing capability. In multiple-input and multiple-output (MIMO) PMCW systems, fast-time code-division multiplexing enables simultaneous transmission from multiple transmitters [...] Read more.

Phase-modulated continuous wave (PMCW) has emerged as a promising waveform candidate for next-generation integrated sensing and communication systems due to its favorable sensing performance and multiplexing capability. In multiple-input and multiple-output (MIMO) PMCW systems, fast-time code-division multiplexing enables simultaneous transmission from multiple transmitters but causes inter-transmitter code interference due to non-ideal cross-correlation properties. The interference is observed to manifest as a low-rank component in the range–Doppler domain while target echoes appear as sparse components. This structural distinction motivates the use of robust principal component analysis (RPCA) for interference mitigation. In practice, conventional RPCA incurs high computational complexity due to the singular value decomposition (SVD) required at every iteration. To address this limitation, we propose an unfolded RPCA network in which each iterative step is mapped to a network stage and SVD is replaced by a factorized low-rank approximation. The proposed network also incorporates stage-wise learnable parameters for adaptive interference mitigation in MIMO PMCW systems. Simulation results demonstrate that the proposed method achieves interference mitigation performance comparable to conventional RPCA with 21.2 times lower inference latency. These results confirm the effectiveness and computational efficiency of the proposed method for real-time mitigation of inter-transmitter code interference in MIMO PMCW systems. Full article

(This article belongs to the Section Radar Sensors)

24 pages, 7825 KB

Open AccessArticle

SY-SLAM: Real-Time Dynamic Indoor RGB-D SLAM with SuperPoint Detection and Asynchronous YOLOv8s-Based Keypoint Suppression

by Shaoshuai Zhi, Shuangfeng Wei, Shan Zhou, Yulan Lao, Mingyang Zhai, Tianyu Yang, Keming Qu and Boyan Jiang

Sensors 2026, 26(11), 3315; https://doi.org/10.3390/s26113315 (registering DOI) - 23 May 2026

Abstract

Traditional visual SLAM pipelines are typically designed under the static-world assumption and often degrade severely in indoor environments with frequent human motion. To improve trajectory accuracy and front-end stability in such scenarios while maintaining real-time throughput, we present SY-SLAM, an RGB-D SLAM system [...] Read more.

Traditional visual SLAM pipelines are typically designed under the static-world assumption and often degrade severely in indoor environments with frequent human motion. To improve trajectory accuracy and front-end stability in such scenarios while maintaining real-time throughput, we present SY-SLAM, an RGB-D SLAM system for dynamic indoor environments with frequent human motion. (S stands for SuperPoint, which is used as a detector-only learned keypoint front-end, and Y stands for YOLO, which provides asynchronous person-aware keypoint suppression based on detected human bounding boxes.) We integrate a TensorRT-deployed detector-only SuperPoint module to improve keypoint repeatability and robustness while retaining ORB binary descriptors for efficient matching and place recognition within the ORB-SLAM3 framework. To avoid feature starvation while preserving keypoint quality, we further introduce an adaptive SuperPoint keypoint selection strategy that applies stricter filtering when keypoints are abundant and relaxes the selection constraints when they are scarce. In parallel, an asynchronous YOLOv8s TensorRT thread performs person detection with temporal bounding-box memory, and keypoints inside detected person regions are removed before ORB descriptor computation and matching to reduce dynamic-feature contamination in the front end. We evaluate SY-SLAM on five dynamic TUM RGB-D fr3 sequences using ATE and RPE metrics. Compared with ORB-SLAM3, SY-SLAM reduces ATE RMSE by 93.45% across four dynamic walking sequences. On the widely reported fr3/w/x sequence, SY-SLAM achieves competitive accuracy with recent dynamic SLAM methods while maintaining real-time performance. The system runs in real time at 46.8 Hz (21.36 ms per frame) on an Intel i9-13900H CPU with an NVIDIA RTX 4070 Laptop GPU. Full article

(This article belongs to the Section Sensors and Robotics)

20 pages, 3635 KB

Open AccessArticle

GPU-Accelerated Signal Processing for Distributed Vibration Sensing Based on OVNA Method

by Alessandro Meoli, Raffaele Vallifuoco, Agnese Coscetta, Luigi Zeni and Aldo Minardo

Sensors 2026, 26(11), 3314; https://doi.org/10.3390/s26113314 (registering DOI) - 23 May 2026

Abstract

Distributed vibration sensing based on optical vector network analysis (OVNA) is a promising technique for measuring dynamic perturbations in optical fibers, but its practical use is limited by the high computational cost of short-time Fourier transform (STFT) and cross-correlation stages. In this work, [...] Read more.

Distributed vibration sensing based on optical vector network analysis (OVNA) is a promising technique for measuring dynamic perturbations in optical fibers, but its practical use is limited by the high computational cost of short-time Fourier transform (STFT) and cross-correlation stages. In this work, we present a GPU-accelerated signal processing pipeline, together with an optimization strategy based on dataflow reduction, mixed-precision arithmetic, and hardware-aware tuning. The proposed implementation reduces the processing time for 200 sweeps from 64.7 s on a single-core CPU to 0.199 s on a modern GPU, while preserving the final shift results, with zero mismatches over 199,199 measurement points. Benchmarking across three GPU generations further shows that STFT benefits more from large on-chip cache resources, whereas cross-correlation scales more closely with memory bandwidth. These results suggest that modern GPUs can significantly reduce the computational burden of OVNA, as well as other distributed sensing methods with a similar processing flow, enabling kHz-rate aggregate throughput from batched processing, supporting real-time-oriented operation on modern GPUs. Full article

(This article belongs to the Special Issue Distributed Sensors: Development and Applications)

16 pages, 1495 KB

Open AccessArticle

DDCATNet: Effective Deep Learning-Based Illumination Color Cast Estimation Approach for Achieving Computational Color Constancy

by Ho-Hyoung Choi

Sensors 2026, 26(11), 3313; https://doi.org/10.3390/s26113313 (registering DOI) - 23 May 2026

Abstract

Digital camera sensors are designed to capture a wide range of incident illuminants, enabling the creation of high-quality images. However, these sensors lack the capability to differentiate between the color of the source illuminant and the actual color (or original color) of the [...] Read more.

Digital camera sensors are designed to capture a wide range of incident illuminants, enabling the creation of high-quality images. However, these sensors lack the capability to differentiate between the color of the source illuminant and the actual color (or original color) of the object being captured. For this reason, the computational color constancy (CCC) was introduced and has been developed over decades. The CCC is an approach to modeling the color perception of the human visual system (HVS) by ensuring accurate object color determination under varying source illuminant conditions. At the core of human visual perception (HVP)-based CCC is attaining higher accuracy in scene illuminant estimation. The emergence of deep convolutional neural networks (DCNNs) was a recent innovation in accurate illuminant estimation, fundamentally transforming the CCC research landscape. Nevertheless, accurate illuminant estimation still remains a huge challenge for both traditional and state-of-the-art (SOTA) approaches. To further advance precision in illuminant estimation, this article presents a novel learning-based illumination color cast estimation approach to HVP-based CCC. Most importantly, the proposed approach is intended to integrate informative features into both channel and spatial regions while preserving long-term dependency feature information with the use of dense skip connections. To achieve these objectives, the proposed Dense Dual Connection Aggregated Transform Network (DDCATNet) architecture is designed to comprise several modules: shallow feature extraction, channel-wise and spatial feature-based Dense Dual Connection (DDC), fusion of the dense channel-wise attention (CA) and spatial attention (SA) branches through a gate mechanism (GM) unit, and aggregate transform. It is worth noting that both the CA blocks and the SA blocks in the DDC module are characterized by dense and cascading connections, meant to preserve long-term feature information and modulate different-level feature information at both global and local scales. The densely connected CA branch (DCA) and the densely connected SA branch (DSA) are also highly effective in securing high-contribution information while suppressing redundant data. The GM unit is integrated at the back of the DDC module, fusing the two DCA and DSA branches to ensure the adaptive merging of useful hierarchical feature information and the extraction of more valuable feature information. As a result, the proposed DDCATNet architecture significantly enhanced precision in illuminant estimation, thereby improving performance. In rigorous experiments on a wide range of datasets, the proposed DDCATNet approach outperformed its SOTA counterparts, validating the efficacy and generalization capabilities, as well as robust camera-invariance, across diverse, single- and multi-illuminant datasets and model architectures. Full article

(This article belongs to the Section Sensing and Imaging)

11 pages, 226 KB

Open AccessArticle

Performance and Tactical Indicators in Elite Formula Kite Sailing Across Different Wind Conditions Using GPS Data

by Javier Riscart-López, Luka Pezelj, Carlos Manuel Pérez-Pérez and Israel Caraballo-Vidal

Sensors 2026, 26(11), 3312; https://doi.org/10.3390/s26113312 (registering DOI) - 23 May 2026

Abstract

The variables determining performance under different wind conditions in Olympic RS:X class sailors have not yet been fully clarified. The aim of this study was to examine the relationship between race performance (final ranking) and GPS-derived variables, including velocity made good (VMG), distance [...] Read more.

The variables determining performance under different wind conditions in Olympic RS:X class sailors have not yet been fully clarified. The aim of this study was to examine the relationship between race performance (final ranking) and GPS-derived variables, including velocity made good (VMG), distance covered, and number of maneuvers, under different wind conditions in Olympic RS:X class sailors. Fifty-four Olympic RS:X class sailors (26 women and 28 men) who competed in a World Cup event were analyzed. Using GPS devices, the first-leg ranking, VMG, number of maneuvers, and distance covered in each segment (upwind, beam reach, and downwind) of every race were evaluated under three wind conditions (light wind: ≥8 to ≤12 knots; moderate wind: >12 to ≤15 knots; strong wind: >15 knots). A total of nine races were analyzed for both the men’s and women’s groups. For both groups and under all wind conditions, first-leg ranking was the variable most strongly associated with overall performance. VMG in upwind and downwind legs was among the most decisive variables in moderate wind conditions for both men and women. In strong wind conditions, VMG during the beam reach was a key variable influencing performance in both groups. In the women’s group, regardless of wind speed, VMG during the first upwind leg was the determining factor for achieving a better first-leg position. In the men’s group, particularly in medium wind conditions, a favorable first-leg position was mainly determined by VMG during the first upwind leg. Full article

(This article belongs to the Special Issue Sports Sensors for Athlete Motion Tracking and Physiological Monitoring: Second Edition)

22 pages, 1748 KB

Open AccessArticle

An Autonomous SAR Image Interpretation Algorithm Based on Multi-Agent Collaborative Scheduling

by Dongdong Lu, Mingjie Zhang, Yibo Guo, Hang Li and Na Liu

Sensors 2026, 26(11), 3311; https://doi.org/10.3390/s26113311 (registering DOI) - 23 May 2026

Abstract

Synthetic Aperture Radar (SAR) image interpretation in dynamic scenarios faces critical challenges, including sluggish multi-agent scheduling responses, sub-optimal task-resource matching, and low full-pipeline collaborative efficiency. To address these issues, this paper proposes an autonomous SAR image interpretation algorithm based on a Mission Control [...] Read more.

Synthetic Aperture Radar (SAR) image interpretation in dynamic scenarios faces critical challenges, including sluggish multi-agent scheduling responses, sub-optimal task-resource matching, and low full-pipeline collaborative efficiency. To address these issues, this paper proposes an autonomous SAR image interpretation algorithm based on a Mission Control Point (MCP)-driven centralized multi-agent collaborative scheduling framework. To address inefficient task–resource matching, a multi-source orchestration model integrating agent states, task characteristics, and environmental dynamics is developed for optimized initial allocation. To mitigate information fragmentation and improve collaboration efficiency across the pipeline, an MCP-based centralized architecture is proposed to achieve unified scheduling and global optimization of multi-stage agents. Furthermore, to enhance adaptability in dynamic environments, a verification-driven adaptive policy continuous optimization mechanism is introduced, allowing the scheduling policy to continuously adapt. Experiments have been conducted on the SARCAP public dataset, and the proposed method achieved a task–agent matching accuracy of 97.98%, an average scheduling latency of 66.1 ms, and a collaborative interpretation speed of 17.9 fps. Compared with MAPPO and conventional centralized scheduling, scheduling efficiency was improved by 12.3% and 18.7%, respectively. Ablation studies further indicate that both the MCP centralized scheduling mechanism and the multi-source information orchestration module significantly contributed to performance, ensuring high accuracy and robustness. Full article

(This article belongs to the Section Environmental Sensing)

15 pages, 3512 KB

Open AccessArticle

A Robust Multi-Branch CNN-LSTM Architecture for Cross-Subject Motor Imagery Classification

by Simone Zini, Federico Bidone and Paolo Napoletano

Sensors 2026, 26(11), 3310; https://doi.org/10.3390/s26113310 (registering DOI) - 23 May 2026

Abstract

Brain–computer interfaces (BCIs) based on motor imagery (MI) aim to convert electroencephalographic (EEG) activity into reliable device commands across users and recording setups. However, low signal-to-noise ratio and strong inter-subject variability still limit true “plug-and-play” deployment without lengthy calibration. To address these challenges, [...] Read more.

Brain–computer interfaces (BCIs) based on motor imagery (MI) aim to convert electroencephalographic (EEG) activity into reliable device commands across users and recording setups. However, low signal-to-noise ratio and strong inter-subject variability still limit true “plug-and-play” deployment without lengthy calibration. To address these challenges, we propose a multi-branch convolutional long short-term memory (CNN-LSTM) architecture that jointly performs multi-scale temporal feature extraction and within-trial sequence modeling. The model employs four parallel 1D convolutional branches with distinct kernel sizes, each followed by an LSTM module and late fusion, combined with group normalization and supervision over sequences of sub-windows within each trial. We evaluate the approach on the EEG Motor Movement/Imagery (EEGMMI) dataset from PhysioNet under strictly subject-independent conditions, and on the ISLab-MI Dataset, a 32-channel wearable-EEG collection designed to assess cross-setup robustness. On EEGMMI, the network achieves up to 82.63% accuracy for binary left/right MI and 74.10% for a four-class task using 4 s trials under 5-fold cross-validation, outperforming an EEGNet-style baseline by 1–10% depending on class count and window length. Under a leave-one-subject-out protocol, the model attains 74.9% mean accuracy for a three-class MI task. Zero-shot transfer to ISLab-MI yields 64.60% and 63.02% accuracy in three- and four-class settings, respectively, while brief subject-specific fine-tuning using only 20% of each session improves performance to 81.38% and 73.48%. These findings show that combining multi-scale convolutional feature extraction with explicit sequence modeling and robust normalization yields accurate, data-efficient, and portable MI decoders suitable for practical BCI applications. Full article

(This article belongs to the Section Biomedical Sensors)

► Show Figures

Figure 1

16 pages, 2180 KB

Open AccessArticle

An M5Stamp Pico-Based IoT Soil Monitoring System for Soil Water–Salinity Diagnosis in a Coastal Reclaimed Pepper Greenhouse

by Leon Nakayama and Ieyasu Tokumoto

Sensors 2026, 26(11), 3309; https://doi.org/10.3390/s26113309 (registering DOI) - 22 May 2026

Abstract

Coastal reclaimed polders with shallow saline groundwater support intensive greenhouse horticulture but require timely diagnosis of root-zone water and salinity conditions. This study developed a compact Internet-of-Things (IoT) monitoring system based on the M5Stamp Pico microcontroller to acquire SDI-12 soil-sensor data, buffer records [...] Read more.

Coastal reclaimed polders with shallow saline groundwater support intensive greenhouse horticulture but require timely diagnosis of root-zone water and salinity conditions. This study developed a compact Internet-of-Things (IoT) monitoring system based on the M5Stamp Pico microcontroller to acquire SDI-12 soil-sensor data, buffer records locally, and transfer them to a low-cost cloud dashboard. Outside-greenhouse validation showed high operational reliability, with a missing observation rate of only 0.9%, and acceptable agreement with a reference TDR100 for both volumetric water content (θ) and bulk electrical conductivity (EC_b). The system was then applied to ridge-position monitoring in a commercial pepper greenhouse on a coastal reclaimed polder. The ridge records captured depth-dependent infiltration and salinity redistribution under drip irrigation, together with contrasting responses between the cultivated layer and shallow groundwater. Potential-based interpretation indicated that the monitored ridge root zone was often not strongly limited by matric potential, whereas osmotic potential derived from pore-water salinity showed reduced water availability even when the soil remained relatively wet. These results demonstrate that continuous real-time monitoring at the ridge position can support diagnosis of root-zone stress and provide useful information for irrigation and fertigation management in salt-affected greenhouse soils. Full article

(This article belongs to the Special Issue Smart Sensors in Precision Agriculture)

30 pages, 5901 KB

Open AccessArticle

Hybrid Analytical and Simulation-Based Approach for Workspace Verification of a Pneumatic Upper Limb Exoskeleton

by Nikita Mayorov, Daniil Teselkin, Denis Dedov and Artem Obukhov

Sensors 2026, 26(11), 3308; https://doi.org/10.3390/s26113308 - 22 May 2026

Abstract

The design of active pneumatic upper limb exoskeletons is complicated by the challenge of reliably determining a kinematically safe workspace. Existing analytical kinematic methods are not sufficient to predict geometric collisions between elements of closed kinematic chains, which poses risks of mechanical damage [...] Read more.

The design of active pneumatic upper limb exoskeletons is complicated by the challenge of reliably determining a kinematically safe workspace. Existing analytical kinematic methods are not sufficient to predict geometric collisions between elements of closed kinematic chains, which poses risks of mechanical damage and threats to user safety during exoskeleton operation. This paper proposes a hybrid algorithm for verifying the workspace of a pneumatic exoskeleton, combining analytical modelling in MATLAB R2020b based on the Product of Exponentials (PoE) method with high-performance static simulation in the Unity environment. At the initial stage, a discrete set comprising 758 million positions of the upper exoskeleton manipulator was generated. Subsequently, a multithreaded two-stage filtering process was implemented: analytical verification of rod stroke limits and angular constraints, followed by the detection of physical intersections of solid-state meshes using the PhysX engine. The results indicate that while the analytical model filters out 99.6% of invalid configurations. Yet, among the remaining positions—formally correct from a mathematical standpoint—up to 50% lead to critical geometric collisions or breaks in the kinematic chain. The computational efficiency of the proposed architecture enabled full static workspace verification in under 20 min. A reachable zone topology was established, revealing pronounced asymmetry and the presence of a “manoeuvrability core” in the user’s anterior hemisphere. The developed algorithm generates a verified set of kinematically safe exoskeleton states, providing a foundation for the kinematic safety layer of a hierarchical control system. These findings demonstrate the necessity of complementing analytical kinematics with physical collision detection when designing hybrid kinematic mechanisms, and the approach can be applied to verify collision-free movement trajectories in various robotic systems. The approach can be applied to verify collision-free movement trajectories in simulation, with physical validation deferred to future work. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

Journal Description

Sensors

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI