Sensors

22 pages, 29295 KB

Open AccessArticle

GeoBoreNet: RGB–Geometry Fusion for Cross-Scene Borehole Detection

by Xuesong Liu, Wenbo Cao, Anke Xu and Emmett J. Ientilucci

Sensors 2026, 26(18), 5963; https://doi.org/10.3390/s26185963 (registering DOI) - 20 Sep 2026

Borehole detection in textured quarry reconstructions is challenging because shadows, rocks, and fractured ground can resemble visible surface openings. We introduce GeoBoreNet, which augments RGB with four pixel-aligned inputs derived from reconstructed surface height: local residual relief, spatial gradient magnitude, spatial Laplacian response, [...] Read more.

Borehole detection in textured quarry reconstructions is challenging because shadows, rocks, and fractured ground can resemble visible surface openings. We introduce GeoBoreNet, which augments RGB with four pixel-aligned inputs derived from reconstructed surface height: local residual relief, spatial gradient magnitude, spatial Laplacian response, and mesh support. Plane removal and resolution-adapted local-background subtraction encode surface morphology, and an expanded first convolution incorporates these channels into standard detectors. Experiments evaluate DINO-R50 and YOLO11x across 11 source scenes using scene-disjoint leave-one-scene-out testing, with five paired training seeds for DINO-R50. In pooled cross-fold evaluation, adding geometry increases AP from

0.483

to

0.548

for DINO-R50 and from

0.385

to

0.450

for YOLO11x, an absolute gain of

0.065

for each detector. Component and local-background scale studies further assess the geometry representation. DINO-R50 source-object-unique AP increases from

0.448

to

0.493

at cross-patch NMS IoU 0.4. In evaluated search regions containing zero-object tiles, AP increases from

0.380

to

0.430

for DINO-R50 and from

0.300

to

0.360

for YOLO11x. These results show that local surface morphology complements RGB for two-dimensional borehole detection in the evaluated scenes. Full article

(This article belongs to the Section Physical Sensors)

► Show Figures

Figure 1

16 pages, 834 KB

Open AccessArticle

FROA-Drive: Failure-Routed Offline Adaptation for Lightweight Vision–Language–Action Autonomous Driving

by Yunhan Xu and Ao Xu

Sensors 2026, 26(18), 5962; https://doi.org/10.3390/s26185962 (registering DOI) - 20 Sep 2026

Abstract

Vision–Language–Action (VLA) models have shown strong potential for end-to-end autonomous driving, yet their post-training commonly relies on expensive simulator interaction or global policy updates. For an already competent pretrained policy, targeted correction is substantially more economical than repeated simulator interaction or global policy [...] Read more.

Vision–Language–Action (VLA) models have shown strong potential for end-to-end autonomous driving, yet their post-training commonly relies on expensive simulator interaction or global policy updates. For an already competent pretrained policy, targeted correction is substantially more economical than repeated simulator interaction or global policy updates. We propose FROA-Drive, a lightweight offline adaptation framework that treats post-training as selective local correction. Starting from the frozen MindDrive-IL 0.5B checkpoint, FROA-Drive constructs a compact policy-centered candidate family around the original Top-1 speed trajectory, learns candidate-relative utility with a 1.51-million-parameter reranker, and uses a validation-calibrated gate to intervene only when an alternative is sufficiently supported. The original meta-actions and path trajectory are preserved by construction, providing an exact identity fallback and preventing unconstrained policy drift. On Chat-B2D-plus-10K, FROA-Drive achieves 72.8% candidate-selection accuracy and 84.2% intervention precision with a 5.4% proxy-violation rate. On Bench2Drive, it reaches a Driving Score of 77.08 and a Success Rate of 52.36%, improving the MindDrive-IL baseline by 1.23 DS and 3.06 percentage points in SR while requiring only 1.51 million trainable parameters and no online rollout during adaptation. These results establish selective local reranking as an effective, computation-efficient post-training strategy for lightweight driving VLA models. Full article

(This article belongs to the Special Issue Intelligent Sensing, Monitoring, and Control of Propulsion Drive Systems for Software-Defined Electric Vehicles)

26 pages, 7494 KB

Open AccessArticle

Learning Motion-Induced Channel Dynamics with Multiresolution Multiplex Graphs for Wi-Fi CSI-Based Human Activity Sensing

by Ying Xiao, Lin Li and Haiyong Zheng

Sensors 2026, 26(18), 5961; https://doi.org/10.3390/s26185961 (registering DOI) - 20 Sep 2026

Abstract

Wi-Fi channel state information (CSI) supports human activity recognition from motion-induced changes in indoor propagation, yet device, environment, and user changes perturb both channel responses and subcarrier relations. We present a model that aligns valid subcarriers across inputs and combines CSI response features [...] Read more.

Wi-Fi channel state information (CSI) supports human activity recognition from motion-induced changes in indoor propagation, yet device, environment, and user changes perturb both channel responses and subcarrier relations. We present a model that aligns valid subcarriers across inputs and combines CSI response features with a multiresolution graph. Temporal decomposition produces residual components at multiple resolutions and a smooth component. The graph propagates information along positive and negative relations between subcarriers within each resolution and uses signal energy to guide propagation across resolutions. A gate incorporates the graph representation into the response classifier. A controlled multipath study further tests the physical interpretation of signed amplitude correlations. Experiments on CSI-Bench yield weighted-F1 scores of 95.96%, 50.16%, 44.08%, and 52.19% on the four protocols, with a cross-domain mean of 48.81%, exceeding the strongest of ten locally reevaluated baselines by 4.84 points. Full article

(This article belongs to the Special Issue Advances in RF Sensing and Signal Processing)

► Show Figures

Figure 1

52 pages, 1061 KB

Open AccessArticle

Blockchain-Backed Revocation and Yang–Baxter Consistency Screening for Zero-Trust IoT Admission Control

by Yair E. Rivera-Julio, Esmeide A. Leal-Narváez and Javier Prieto Tejedor

Sensors 2026, 26(18), 5960; https://doi.org/10.3390/s26185960 (registering DOI) - 20 Sep 2026

Abstract

IoT deployments handle credential hygiene reactively: cloned, replayed, or stale credentials are typically discovered only after misuse, and revocation state is often propagated through centralized lists whose integrity cannot be independently verified. This article introduces the Yang–Baxter IoT Consistency Gateway (YB-IoT-CG), a Zero-Trust [...] Read more.

IoT deployments handle credential hygiene reactively: cloned, replayed, or stale credentials are typically discovered only after misuse, and revocation state is often propagated through centralized lists whose integrity cannot be independently verified. This article introduces the Yang–Baxter IoT Consistency Gateway (YB-IoT-CG), a Zero-Trust admission-control framework that pushes an algebraic layer of credential screening to the edge and anchors security evidence on a modeled permissioned-ledger architecture. YB-IoT-CG operates after conventional secret-based authentication and Yang–Baxter equality is used as an execution-consistency invariant rather than as proof of device identity or message authenticity. Each authenticated message is hashed into a digest and reduced to an algebraic transcript that is verified over two Yang–Baxter traversal paths. Beyond equality checking, chain-proximity metrics inspired by time–memory trade-off analysis, nonce and timestamp freshness heuristics, and contextual risk scoring identify credentials that should be proactively challenged or revoked before telemetry is admitted. The proactive risk component is deterministic and policy-based rather than a learned predictive model. Decisions are batched into Merkle trees whose roots are represented through a permissioned-ledger model, credential revocation is propagated through a modeled on-chain registry, and device identities are bound to W3C Decentralized Identifiers (DIDs) with verifiable credentials. This design provides tamper-evident audit support while keeping ledger interaction off the packet decision path. Validation is based on a controlled Python 3.13.7 packet-level simulation and a parameterized ledger model, not on a physical IoT or live Hyperledger Fabric deployment. Across 60 seeded simulation runs, the complete post-authentication screening pipeline obtained a median incremental decision time of 7.8

μ

s, 99.4% aggregate decision accuracy for the modeled attack classes, a 0.43% false rejection rate, and a 31.4 ms amortized ledger service-time equivalent per decision for a batch size of 64. The 99.4% value is a property of the composed freshness/context/registry/YB policy and is not a YB-only detection rate; the YB-specific positive guarantee is limited to the isolated fault class of Proposition 7. These results provide simulation-based evidence supporting a lightweight, explainable, auditable, and proactive approach to credential hygiene at the edge. Full article

(This article belongs to the Special Issue Feature Papers in Smart Sensing and Intelligent Sensors 2026)

► Show Figures

Figure 1

20 pages, 928 KB

Open AccessArticle

A Hybrid Frequency Estimation Framework for Long-Range FMCW LiDAR Under Low-SNR Conditions

by Yating Fang, Xiaohai Yu, Chaochao Zhang, Jia He, Yuanying Zhang, Hua Zheng and Xiaoqin Zhu

Sensors 2026, 26(18), 5959; https://doi.org/10.3390/s26185959 (registering DOI) - 20 Sep 2026

Abstract

Frequency-modulated continuous-wave (FMCW) light detection and ranging (LiDAR) has become increasingly significant for long-distance ranging fields. However, the beat signal is vulnerable to noise interference during detection, resulting in a low signal-to-noise ratio (SNR) and inaccurate ranging results. Although traditional methods have improved [...] Read more.

Frequency-modulated continuous-wave (FMCW) light detection and ranging (LiDAR) has become increasingly significant for long-distance ranging fields. However, the beat signal is vulnerable to noise interference during detection, resulting in a low signal-to-noise ratio (SNR) and inaccurate ranging results. Although traditional methods have improved accuracy through various denoising and spectral analysis algorithms, they often struggle to maintain robust performance in long-range scenarios under low-SNR conditions. To address this challenge, this paper proposes the HOIGL, a hybrid frequency estimation algorithm which integrates Orthogonal Matching Pursuit (OMP) for initial value guidance and Gray Wolf Optimization (GWO) for local search. Specifically, by leveraging the frequency-domain sparsity of the signal, OMP performs piecewise sparse representation to yield a coarse initial frequency, within which GWO conducts a refined continuous local search to overcome dictionary discretization bias based on a custom fitness function. Monte Carlo simulations in MATLAB demonstrate that the proposed algorithm achieves an RMSE of 2.63 m compared to over 84.34 m for other methods at −20 dB SNR, while maintaining a low and stable processing time of around 0.498 s. Finally, fiber-optic-link experimental results verify the performance of HOIGL for hundred-meter-scale range detection. Full article

(This article belongs to the Special Issue Progress in LiDAR Technologies and Applications)

24 pages, 5550 KB

Open AccessArticle

Parameter Estimation of Non-Cooperative Composite Signals Based on Joint Time-Polarization Domain Separation

by Yun Zhou, Fulai Wang, Hao Wu, Chen Pang, Yongzhen Li and Ping Wang

Sensors 2026, 26(18), 5958; https://doi.org/10.3390/s26185958 (registering DOI) - 20 Sep 2026

Abstract

The detection and identification process in radar is susceptible to disturbances from numerous non-cooperative jamming signals in addition to target signals. Accurately estimating the parameters of these non-cooperative signals is crucial for extracting the target signal. Addressing the current lack of effective solutions [...] Read more.

The detection and identification process in radar is susceptible to disturbances from numerous non-cooperative jamming signals in addition to target signals. Accurately estimating the parameters of these non-cooperative signals is crucial for extracting the target signal. Addressing the current lack of effective solutions for estimating parameters of different types of uncooperative combined signals, this paper proposes a method for parameter estimation of non-cooperative composite signals based on joint time-polarization domain separation. This method combines amplitude and phase information from both the time and polarization domains of non-cooperative signals and integrates the JADE separation algorithm to achieve the separation and estimation of specific non-cooperative signal combinations. Experiments compare the separation performance of this method with other signal separation algorithms for non-cooperative jamming signal combinations. The results indicate that, at jamming-to-noise ratio of −10 dB to 15 dB, this method achieves better separation accuracy and precision for specific non-cooperative jamming combinations, while also accurately estimating the parameters of the non-cooperative jamming signals. Full article

(This article belongs to the Topic Radar Signal and Data Processing with Applications, 2nd Edition)

35 pages, 8385 KB

Open AccessArticle

Design, Development, and Laboratory Validation of a Low-Cost Multi-Sensor System for Bridge Structural Health Monitoring with Scour-Related Environmental Sensing

by Matilde Bidone, Mauro Aimar, Marco Civera, Alessio Carullo and Alberto Vallan

Sensors 2026, 26(18), 5957; https://doi.org/10.3390/s26185957 (registering DOI) - 20 Sep 2026

Abstract

This paper reports the design, development and laboratory validation of a low-cost multi-sensor system for bridge Structural Health Monitoring with complementary scour-related environmental sensing. The proposed sensing apparatus is intended for continuous dynamic (vibration-based) and quasi-static structural monitoring, in particular for bridge applications. [...] Read more.

This paper reports the design, development and laboratory validation of a low-cost multi-sensor system for bridge Structural Health Monitoring with complementary scour-related environmental sensing. The proposed sensing apparatus is intended for continuous dynamic (vibration-based) and quasi-static structural monitoring, in particular for bridge applications. It employs a distributed wired architecture comprising two custom printed circuit boards, BridgeWatch and EnvironMonitor. BridgeWatch nodes integrate MEMS accelerometers and gyroscopes, enabling acceleration measurements, quasi-static inclination estimates from gravity components, angular-rate measurements and structural surface-temperature acquisition. The EnvironMonitor node acquires wind, rainfall, air temperature, humidity and bridge water level; these variables provide environmental and hydrological context relevant to scour risk, without directly measuring riverbed elevation or scour depth. A Raspberry Pi coordinates acquisition, synchronisation and communication through RS485 and RS232 links. The hardware was implemented through custom PCB design and dedicated firmware. The principal contribution is the system-level integration of structural and environmental measurements in one open architecture. Laboratory validation comprised communication and synchronisation checks, shaker tests, a scaled structural benchmark, climatic-cell tests and environmental-input simulations. The results demonstrate functional acquisition and relevant dynamic-feature extraction under controlled conditions. These outcomes enable future full-scale in situ validation and testing on full-size case studies. Full article

(This article belongs to the Special Issue Recent Advances in Structural Health Monitoring of Bridges)

► Show Figures

Graphical abstract

24 pages, 4426 KB

Open AccessArticle

A Therapist-Configurable Serious Game System for Upper-Limb Neurorehabilitation: A Pilot Feasibility Study

by Juan J. Sánchez-Gil, Aurora Sáez, Juan José Ochoa-Sepúlveda, Rafael López-Luque, Laura Muñoz-Millán and Eduardo Cañete-Carmona

Sensors 2026, 26(18), 5956; https://doi.org/10.3390/s26185956 (registering DOI) - 20 Sep 2026

Abstract

Upper-limb neurorehabilitation requires adaptable practice. This pilot feasibility study evaluated a therapist-configurable serious-game system in which a handle equipped with an RGB sensor, embedded processing, and wireless communication converted detections of reconfigurable colored targets into control events for 17 games. Thirteen neurological patients [...] Read more.

Upper-limb neurorehabilitation requires adaptable practice. This pilot feasibility study evaluated a therapist-configurable serious-game system in which a handle equipped with an RGB sensor, embedded processing, and wireless communication converted detections of reconfigurable colored targets into control events for 17 games. Thirteen neurological patients began six supervised sessions over three weeks alongside usual care; 11 completed them (84.6%). Including one session completed by each participant who withdrew, 68 sessions were delivered overall (approximately 34 h of scheduled time). From a sensing-operability perspective, no session ended prematurely because of a sensing or other technical failure; RGB recalibration was required in 4 sessions (5.9%). Measures included session logs, System Usability Scale, Intrinsic Motivation Inventory, In-Game Game Experience Questionnaire, Pittsburgh Rehabilitation Participation Scale, Box and Block Test, and exploratory movement analysis. Mean SUS was 76.1 ± 12.6; IMI interest/enjoyment and perceived competence were 6.47/7 and 6.39/7; exploratory post-activity iGEQ positive affect was 3.64/4; and PRPS means ranged from 4.64 to 5.45/6. BBT scores increased descriptively in 11 of 12 assessed affected limbs. The system was feasible and acceptable as a supervised adjunct, but exploratory findings do not establish clinical efficacy. Full article

(This article belongs to the Special Issue Innovative IoT Devices in Biomedical Monitoring: From Design to Application)

► Show Figures

Figure 1

53 pages, 1832 KB

Open AccessArticle

Bias-Aware Detection Limits, Calibration Transfer and Matched-Control Robustness Assessment in Dual-Parameter Photonic Refractive-Index and Temperature Sensing: A Coupled Interface-Mode Multilayer Case Study

by Agah Oktay Ertay and Muhammed Mustafa Ertay

Sensors 2026, 26(18), 5955; https://doi.org/10.3390/s26185955 (registering DOI) - 20 Sep 2026

Abstract

Dual-parameter photonic sensors are usually reported through a nominal sensitivity and one detection limit, without stating which statistic that limit is or whether it survives transfer between devices. This computational study supplies that evaluation for one structure, a 36-layer one-dimensional multilayer read in [...] Read more.

Dual-parameter photonic sensors are usually reported through a nominal sensitivity and one detection limit, without stating which statistic that limit is or whether it survives transfer between devices. This computational study supplies that evaluation for one structure, a 36-layer one-dimensional multilayer read in transmission, whose Zak-phase-distinct TiO

_{2}

/SiO

_{2}

photonic-crystal sections enclose a 600 nm analyte cavity and a 500 nm thermo-optic reference cavity, each carrying a 5 nm ITO/5 nm TiO

_{2}

nanolaminate insert. Two coupled interface resonances at 1517 and 1651 nm, with loaded Q of 232 and 208 and refractive-index (RI) sensitivities of 90.71 and 329.41 nm/RIU, are inverted by a bounded nonlinear calibration to

2.43 \times 10^{- 5}

RIU and

0.155

^{\circ} C

; the temperature channel reports the device temperature. Probability-of-detection limits at 1% false alarm and 95% detection are

4.36 \times 10^{- 5}

RIU and

0.123

^{\circ} C

; they are set by the calibration standards and the wavelength reference, not by the linewidth. Transferring one calibration between devices worsens them 81-fold and 203-fold; a three-point per-device correction removes 84–93% of that loss. A trivial control matched on wavelength, Q, transmission, thickness and RI sensitivity shows no topological robustness advantage. Applied to the design itself, the same evaluation shows that the modes are cavity-selected, that hyperbolicity brings no benefit, and that the nanolaminate-free stack is preferred. Full article

(This article belongs to the Special Issue Feature Papers in Optical Sensors 2026)

24 pages, 7960 KB

Open AccessArticle

Awareness-Guided Self-Supervised Visual Odometry via Fusion of Pseudo-Depth and Residual-Flow Cues

by Shi Zhou, Zijun Yang, Zhen Li, Xianchang Li, Yuchen Sun and Lifeng Zhang

Sensors 2026, 26(18), 5954; https://doi.org/10.3390/s26185954 (registering DOI) - 20 Sep 2026

Abstract

Self-supervised monocular visual odometry (VO) commonly relies on photometric consistency and rigid-scene assumptions to learn camera motion, making pose estimation vulnerable to unreliable cross-frame observations caused by limited geometric observability and independently moving objects. To address this problem, we propose an awareness-guided self-supervised [...] Read more.

Self-supervised monocular visual odometry (VO) commonly relies on photometric consistency and rigid-scene assumptions to learn camera motion, making pose estimation vulnerable to unreliable cross-frame observations caused by limited geometric observability and independently moving objects. To address this problem, we propose an awareness-guided self-supervised visual odometry framework that emphasizes reliable cross-frame observations for pose learning. Pseudo-depth is exploited to characterize geometric observability and identify regions that are unreliable for cross-frame reconstruction, while residual flow is employed to capture motion inconsistency introduced by independently moving objects. The resulting observable static regions are incorporated into pose estimation at both the input and loss levels, enabling unreliable observations to be suppressed before feature extraction and during optimization. Qualitative experiments on KITTI and Cityscapes verify the effectiveness of pseudo-depth and residual flow in extracting geometric observability and dynamic-motion information across different driving scenes. Quantitative experiments on the KITTI Odometry dataset show that the proposed method achieves average relative translation and rotation errors of 6.04% and 2.42

^{\circ}

/100 m, together with an ATE of 0.88 m and an ARE of 0.20

^{\circ}

, demonstrating improved ego-motion and trajectory estimation. Full article

(This article belongs to the Section Optical Sensors)

19 pages, 3761 KB

Open AccessArticle

Multi-Granular Embedding and Hybrid Encoder–Decoder Architecture with Temporal Linear Regression Bypass for Sensor-Based Time Series Forecasting in Smart Infrastructure

by Hsu-Yung Cheng, Huan-Hsuan Lin, Chi-Lun Jiang and Chih-Chang Yu

Sensors 2026, 26(18), 5953; https://doi.org/10.3390/s26185953 (registering DOI) - 20 Sep 2026

Abstract

In this work, we propose a robust deep encoder–decoder neural network that integrates multiple deep sequence models for time series forecasting in sensor-based applications. The proposed approach features a sophisticated input embedding layer that integrates scalar features, position encodings, and multi-granularity time embeddings. [...] Read more.

In this work, we propose a robust deep encoder–decoder neural network that integrates multiple deep sequence models for time series forecasting in sensor-based applications. The proposed approach features a sophisticated input embedding layer that integrates scalar features, position encodings, and multi-granularity time embeddings. To improve computational efficiency, we employ a generative-style decoder inspired by the Informer architecture, which enables full-length sequence prediction in a single iteration, avoiding the latency associated with traditional autoregressive models. The architecture is highly flexible, supporting either Gated Recurrent Unit (GRU) or Transformer blocks as foundational layers to process latent temporal dependencies. Recognizing that purely nonlinear deep learning models often struggle to capture sharp, short-term local variations due to inherent input scale insensitivity, we augment our model with a parallel temporal linear regression unit. This bypass mechanism captures linear local trends, effectively combining the predictive power of nonlinear deep feature extraction with the sensitivity of linear regression. Experimental evaluations on the Electricity Consumption (EC) and Speed Index California (SIC) datasets demonstrate that the GRU-based implementation reduces the average Mean Squared Error (MSE) by approximately 22% and 16%, respectively, compared with the best-performing baseline across different output lengths. Ablation studies further confirm that the temporal linear regression bypass substantially improves forecasting accuracy, particularly on the EC dataset, where it provides an average MSE reduction of approximately 55%. Full article

(This article belongs to the Special Issue Smart Sensing and Embedded AI for IoT Systems)

► Show Figures

Figure 1

1 pages, 127 KB

Open AccessRetraction

RETRACTED: Prażnowski et al. Assessment of the Road Surface Condition with Longitudinal Acceleration Signal of the Car Body. Sensors 2020, 20, 5987

by Krzysztof Prażnowski, Jarosław Mamala, Michał Śmieja and Mariusz Kupina

Sensors 2026, 26(18), 5952; https://doi.org/10.3390/s26185952 (registering DOI) - 20 Sep 2026

Abstract

The journal retracts the article titled “Assessment of the Road Surface Condition with Longitudinal Acceleration Signal of the Car Body” [...] Full article

15 pages, 6338 KB

Open AccessArticle

Subject-Specific Deep Learning Model for Motor Imagery Direction Decoding

by Praveen K. Parashiva, Sagila Gangadharan Kutteri and A. P. Vinod

Sensors 2026, 26(18), 5951; https://doi.org/10.3390/s26185951 (registering DOI) - 20 Sep 2026

Abstract

Hemispheric strokes impair motor control in contralateral body parts, necessitating effective rehabilitation strategies. Motor imagery-based brain–computer interfaces (MI-BCIs) promote neuroplasticity, aiding the recovery of motor functions. While deep learning has shown promise in decoding MI actions for stroke rehabilitation, existing studies largely focus [...] Read more.

Hemispheric strokes impair motor control in contralateral body parts, necessitating effective rehabilitation strategies. Motor imagery-based brain–computer interfaces (MI-BCIs) promote neuroplasticity, aiding the recovery of motor functions. While deep learning has shown promise in decoding MI actions for stroke rehabilitation, existing studies largely focus on bilateral MI actions and are limited to offline evaluations. Decoding directional information from unilateral MI, however, offers a more natural control interface with greater degrees of freedom but remains challenging due to spatially overlapping neural activity. This work proposes a novel deep learning framework for online decoding of binary directional MI signals from the dominant hand of 20 healthy subjects. The proposed method employs EEGNet-based convolutional filters to extract temporal and spatial features. The EEGNet model is enhanced by squeeze-and-excitation (SE) layers that rank the electrode importance and feature maps. A subject-independent model is initially trained using calibration data from multiple subjects and fine-tuned for subject-specific adaptation. The performance of the proposed method is evaluated using subject-specific online session data. The proposed method achieved an average right vs. left binary direction-decoding accuracy of

58.7 \pm 8 %

for unilateral MI tasks, outperforming the existing deep learning models. Additionally, the SE-layer ranking offers insights into electrode contribution, enabling potential subject-specific BCI optimization. The findings highlight the efficacy of the proposed method in advancing MI-BCI applications for a more natural and effective control of BCI systems. Full article

(This article belongs to the Special Issue EEG Signal Processing Techniques and Applications—3rd Edition)

► Show Figures

Figure 1

35 pages, 10068 KB

Open AccessArticle

A Mesh Updating Framework for Shell-Element Finite Element Models Based on Projection of Anomaly Regions Extracted from Point-Cloud Data

by Jiexiu Wang and Mayuko Nishio

Sensors 2026, 26(18), 5950; https://doi.org/10.3390/s26185950 (registering DOI) - 20 Sep 2026

Abstract

Finite element (FE) analysis that incorporates structural damages extracted using computer vision (CV) methods based on three-dimensional point-cloud data (PCD) is effective for evaluating the residual capacity of structures. In this study, observable surface damages are regarded as surface anomaly regions on the [...] Read more.

Finite element (FE) analysis that incorporates structural damages extracted using computer vision (CV) methods based on three-dimensional point-cloud data (PCD) is effective for evaluating the residual capacity of structures. In this study, observable surface damages are regarded as surface anomaly regions on the structure. Based on this assumption, a modular mesh-model updating framework is proposed to incorporate damage information into shell-element FE models through the anomaly detection from point-cloud data. The framework comprises three modules: a perception module for the anomaly detection and anomaly-region localization, a description module for the anomaly-region quantification, and a remeshing module for mapping the anomaly region onto the FE model. Its performance was evaluated using a steel angle member specimen with artificially introduced anomaly regions representing damage. The updating results were evaluated in terms of both geometric accuracy and FE analysis (FEA) applicability by comparison with the reference FE model manually constructed from the designed anomaly-region geometries. The Intersection over Union (IoU) between corresponding anomaly regions in the updated and reference models ranges from 0.44 to 0.87, including the individual plate results for cross-surface cases. Through static elastoplastic analysis, the predicted reductions in ultimate load-bearing capacity differed by approximately 1.2 percentage points between the two updated models. Moreover, local stress redistribution showed qualitative agreement, although quantitative discrepancies remained in the magnitudes and locations of stress concentrations. Together with several supporting components, the proposed framework provides an effective workflow for integrating PCD-based CV techniques into FE model updating and has potential for FEA-based damage assessment of structures. Full article

(This article belongs to the Section Sensing and Imaging)

15 pages, 15228 KB

Open AccessArticle

A Stand-Alone Wearable Measurement System for Continuous Hand Tremor Monitoring

by Laura Arruzzoli, Carmen Galletta, Giovanni Gugliandolo, Mariangela Latino, Giovanni Crupi, Cristiano De Marchis and Nicola Donato

Sensors 2026, 26(18), 5949; https://doi.org/10.3390/s26185949 (registering DOI) - 20 Sep 2026

Abstract

Progressive neurodegenerative diseases are often associated with motor symptoms, among which tremor is one of the most debilitating manifestations, significantly compromising daily activities and quality of life. Continuous, non-invasive monitoring of tremor characteristics helps to assess disease progression and optimize treatment strategies. For [...] Read more.

Progressive neurodegenerative diseases are often associated with motor symptoms, among which tremor is one of the most debilitating manifestations, significantly compromising daily activities and quality of life. Continuous, non-invasive monitoring of tremor characteristics helps to assess disease progression and optimize treatment strategies. For this reason, wearable measurement systems can provide an objective and non-invasive means for the quantitative characterization and monitoring of tremor-related movements. This study addresses the development of a compact, low-power wearable board for real-time tremor monitoring based on the Nicla Sense ME embedded system. The proposed device performs on-board processing to extract spectral features of movement and wirelessly sends them via Bluetooth Low Energy to a remote host, enabling on-demand visualization or storage of tremor data. The measurement system was technically validated through a comparative experimental protocol based on the simultaneous acquisition of tremor signals using the proposed device and a commercial Shimmer3 IMU as the reference system. The comparison considered both the similarity of the measured spectrograms and the agreement in dominant-frequency estimation. Spectrogram correlation coefficients reached 0.85, with amplitude MAE and RMSE values as low as 0.07 m/s² and 0.14 m/s², respectively. For dominant-frequency estimation, Pearson correlation coefficients reached 0.99, with MAE and RMSE values as low as 0.17 Hz and 0.20 Hz. These results support the feasibility of the proposed architecture as a practical wearable solution for quantitative hand tremor monitoring. Full article

(This article belongs to the Special Issue Wearable Sensors for Gait, Human Motion and Health Monitoring)

► Show Figures

Figure 1

26 pages, 8842 KB

Open AccessArticle

TGLDA: Tail-Guided Low-Density Normal Data Augmentation for Industrial Control Systems

by JuHyeon Lee, Ilhwan Ji, Seungho Jeon and Jung Taek Seo

Sensors 2026, 26(18), 5948; https://doi.org/10.3390/s26185948 (registering DOI) - 20 Sep 2026

Abstract

Anomaly detection systems for detecting cyber threats in industrial control systems (ICSs) can suffer performance degradation because rare normal samples may not be sufficiently learned. To address this problem, various data augmentation methods have been studied. However, existing methods may generate data in [...] Read more.

Anomaly detection systems for detecting cyber threats in industrial control systems (ICSs) can suffer performance degradation because rare normal samples may not be sufficiently learned. To address this problem, various data augmentation methods have been studied. However, existing methods may generate data in unknown regions based on model decision boundaries or unnecessarily augment regions that already contain sufficient samples. Many existing methods also require anomalous data. To address these limitations, we propose Tail-Guided Low-Density Normal Data Augmentation (TGLDA), which operates using only normal data. TGLDA consists of a low-density region estimator and a data generator. It identifies low-density regions with relatively few samples in the observed data distribution and selectively augments these regions. TGLDA uses feature-tail seeds as heuristic starting points for generating samples in low-density regions. Experiments on a water treatment cybersecurity dataset show that TGLDA successfully generates data focused on low-density regions compared with other data augmentation methods. When each augmentation method was applied to baseline anomaly detection models, TGLDA achieved the highest Accuracy and F1-score across all models. These results show that TGLDA can mitigate problems caused by a lack of rare normal samples in real-world ICS environments where only normal data are available. Full article

(This article belongs to the Topic AI and Data-Driven Advancements in Industry 4.0, 2nd Edition)

22 pages, 5384 KB

Open AccessArticle

Temperature-Dependent Electromechanical and Acoustic Responses of a Tubular PZT-5A Transducer: Finite-Element Modeling and Experimental Investigation

by Ping Han, Baohai Tan, Kai Zhang, Yuanda Su and Zhicheng Zheng

Sensors 2026, 26(18), 5947; https://doi.org/10.3390/s26185947 (registering DOI) - 20 Sep 2026

Abstract

Temperature-induced variations in piezoelectric material properties can detune acoustic-logging transducers and alter their electrical, mechanical, and acoustic responses. This study establishes a temperature-dependent finite-element model of a tubular lead zirconate titanate (PZT-5A) transducer over 0–200 °C. Temperature-dependent dielectric, piezoelectric, and elastic parameters were [...] Read more.

Temperature-induced variations in piezoelectric material properties can detune acoustic-logging transducers and alter their electrical, mechanical, and acoustic responses. This study establishes a temperature-dependent finite-element model of a tubular lead zirconate titanate (PZT-5A) transducer over 0–200 °C. Temperature-dependent dielectric, piezoelectric, and elastic parameters were implemented as constitutive functions. Frequency-domain impedance simulations were compared with impedance measurements over 20–200 °C, while transient fluid-loaded simulations and laser-vibrometry measurements were used to characterize the acoustic-pressure and vessel-wall vibration responses. The simulated resonant frequency decreased from 21.379 to 19.35 kHz, and the measured value decreased from 21.135 to 19.55 kHz. The simulated static capacitance increased from 18.8 to 81.0 nF, whereas the measured value increased from 23.7 to 93.0 nF. An isolated geometric-expansion sensitivity analysis indicated that omitted dimensional changes may partly account for this discrepancy. The fitted motional resistance decreased in both the simulations and experiments. The vessel-wall vibration-velocity squared integral decreased from 0.03483 to

0.0067 {mm}^{2}

/s, and the simulated acoustic-pressure squared integral decreased from 2.14926 to 1.59839 Pa²·s. The frequency corresponding to the maximum sound-pressure spectrum also decreased from 22.644 to 21.607 kHz. The agreement in the principal temperature-dependent trends indicates that the model captures the dominant electromechanical behavior, while the remaining capacitance discrepancy highlights the influence of thermal deformation and other nonideal experimental factors. The proposed framework provides a basis for performance prediction and temperature compensation of piezoelectric transducers used in acoustic logging and other elevated-temperature acoustic systems. Full article

(This article belongs to the Collection Ultrasound Transducers)

► Show Figures

Figure 1

44 pages, 35095 KB

Open AccessArticle

Performance Assessment of the Earthview BluBird Continuous Emission Monitoring System for Oil and Gas Facility Emissions Monitoring

by James Maslanik, Kevin Yoon and Frederick Givhan

Sensors 2026, 26(18), 5946; https://doi.org/10.3390/s26185946 (registering DOI) - 20 Sep 2026

Abstract

The Earthview BluBird continuous methane emissions monitoring system participated in single-blind controlled-release testing in 2022 and 2024, with the goal of evaluating the leak detection and quantification capabilities of these metal-oxide point sensors under realistic field conditions. The METEC ADED test program consisted [...] Read more.

The Earthview BluBird continuous methane emissions monitoring system participated in single-blind controlled-release testing in 2022 and 2024, with the goal of evaluating the leak detection and quantification capabilities of these metal-oxide point sensors under realistic field conditions. The METEC ADED test program consisted of natural-gas releases (“experiments”) at a simulated gas production facility, with each experiment comprising up to five simultaneous releases from different locations. Here, we analyze the results according to the detection of at least one release per experiment (“per experiment” detection) and the detection of all releases. Using METEC’s standard grading criteria, 91% of ADED 2024 experiments were detected; a substantial improvement compared to the 2022 results. Undetected releases during multi-release experiments, rather than undetected experiments, account for the large majority of missed releases. Detection correlated more strongly with release duration than with release rate. Detection rates and probability of detection improve when alternative classification criteria are applied. Sources were correctly identified at the equipment group level for 75% of single-release experiments. Actual and estimated methane-emission mass summed over daily to monthly intervals correlate significantly, with BluBird underestimating mass by 36% for single-release experiments. This underestimate is dominated by emission rate error rather than release detection capability. Full article

(This article belongs to the Special Issue Sensor-Based Systems for Environmental Monitoring and Assessment)

► Show Figures

Figure 1

31 pages, 3787 KB

Open AccessArticle

Digital Predistortion of Wideband Power Amplifiers Using Functionally Decoupled Envelope-Assisted Attention-Guided Recurrent Architecture

by Bingwen Qiu, Xiaoyu Li and Yunjie Zhao

Sensors 2026, 26(18), 5945; https://doi.org/10.3390/s26185945 (registering DOI) - 19 Sep 2026

Abstract

Wideband power amplifiers (PAs) operating with high-order modulation signals exhibit strong nonlinear distortion and dynamic memory effects, making real-time digital predistortion (DPD) increasingly challenging under strict computational constraints. This work proposes a functionally decoupled neural DPD architecture, termed EA-CCF-AttGRU, which explicitly separates instantaneous [...] Read more.

Wideband power amplifiers (PAs) operating with high-order modulation signals exhibit strong nonlinear distortion and dynamic memory effects, making real-time digital predistortion (DPD) increasingly challenging under strict computational constraints. This work proposes a functionally decoupled neural DPD architecture, termed EA-CCF-AttGRU, which explicitly separates instantaneous nonlinear feature representation from temporal memory compensation within a unified end-to-end framework. Instead of introducing envelope features, cross-channel fusion, and recurrent attention as isolated modules, the proposed architecture assigns different compensation functions to dedicated components: envelope-assisted augmentation and point-wise cross-channel fusion enhance instantaneous nonlinear representation, while attention-guided recurrent modeling captures dynamic memory effects. A global linear bypass further reduces the burden of nonlinear compensation by preserving the linear transformation. Experimental results under a 160 MHz 1024-ary quadrature amplitude modulation (1024-QAM) baseband excitation with a 10.38 dB peak-to-average power ratio (PAPR) demonstrate that the proposed method achieves an adjacent channel leakage ratio (ACLR) of −65.91 dBc, a normalized mean square error (NMSE) of −57.84 dB, and an error vector magnitude (EVM) of 0.07% with only 6009 trainable parameters. The proposed architecture achieves an effective complexity–performance trade-off for wideband DPD applications and provides potential for future hardware-oriented implementation and synthesis validation. Full article

(This article belongs to the Special Issue Microwave, RF, and mm-Wave Circuits for Advanced Wireless Communications and Sensing)

► 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