IEEE Sensors Journal最新文献

{"title":"Design and Fabrication of Embedded Contact Pressure Sensor for Orthotic Interfaces via Multimaterial 3-D Printing","authors":"Xinyuan Song;Xinyang Tan;Fan Feng;Quan Li;Hongwei Zhang;Lifeng Lao","doi":"10.1109/JSEN.2026.3712313","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3712313","url":null,"abstract":"Built-in contact pressure sensing at orthotic interfaces is important for monitoring excessive pressure, assessing wearing comfort, and enabling individualized adjustment. However, existing sensorized orthoses mainly rely on postattached or externally mounted sensors, which often exhibit limited conformity and weak structural integration on complex curved surfaces. Therefore, this study proposes a design and fabrication method for sensorized orthoses based on multimaterial 3-D printing, in which the sensor units are directly integrated into the orthosis during fabrication. A 3-D-printed piezoresistive sensor for orthotic applications was designed, and the effects of the infill patterns (triangle, honeycomb, and gyroid) and densities (25%–45%) of its piezoresistive component on sensing performance were evaluated through multiphysics finite element simulations and experimental testing. The results showed that, within the pressure range of 0–0.25 MPa, the honeycomb pattern with 40% infill density exhibited the optimal overall performance, with high sensitivity (0.43 <inline-formula> <tex-math>${mathrm {MPa}}^{-{1}}$ </tex-math></inline-formula>), good linearity (error = 5.60%), high accuracy (error <5%), low hysteresis error (<4%), excellent within-sample repeatability (error <3%), and high cross-sample repeatability (ICC = 0.92). To validate the proposed method, a sensorized hand orthosis with seven built-in sensors was developed to measure contact pressures at key anatomical locations and differentiate pressure distribution patterns across different gesture conditions. This method provides high design flexibility, streamlines the fabrication of sensorized orthoses, ensures precise anatomical adaptation to complex body interfaces, and shows potential for extension to other orthotic and assistive devices.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"25483-25495"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bandwidth Extension and Wideband Noise Suppression for Mode-Matched Gyroscopes Based on Second Pair of Pole-Zero Configuration","authors":"Yukai Huang;Pu Chen;Shenhu Huang;Zhuolin Yu;Qilong Wu;Xiang Lv;Yi Zhou;Zhaorong Ke;Bo Jiang;Yan Su","doi":"10.1109/JSEN.2026.3713487","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3713487","url":null,"abstract":"Mode-matching effectively enhances the signal-to-noise ratio of MEMS gyroscopes but considerably narrows the mechanical bandwidth. Force-to-rebalance (FTR) closed-loop control extends the operational bandwidth yet is constrained by the system phase margin (PM). Traditional pole-zero cancellation compensation enables bandwidth extension while maintaining PM, but severely deteriorates noise performance, leaving the bandwidth–noise tradeoff unresolved. This article first analytically characterizes the noise behavior of the single-pair pole-zero compensated FTR system, identifying that channel gain is the key factor of this trade-off. A new strategy is then proposed that exploits the previously overlooked interaction between the compensator pole and the proportional-integral (PI) zero, introducing the frequency ratio <inline-formula> <tex-math>$uplambda $ </tex-math></inline-formula> as an explicit design variable. Analysis reveals that configuring <inline-formula> <tex-math>$uplambda gt 1$ </tex-math></inline-formula> reduces the channel gain required to achieve the target bandwidth, and a quantitative selection criterion for <inline-formula> <tex-math>$uplambda $ </tex-math></inline-formula> is established under simultaneous bandwidth, stability, and noise constraints. Following the proposed selection criterion, experimental results on the same hardware platform demonstrate that at 100-Hz closed-loop bandwidth, angle random walk (ARW) is improved from 0.0789°/<inline-formula> <tex-math>$!surd $ </tex-math></inline-formula>h to 0.0186°/<inline-formula> <tex-math>$!surd $ </tex-math></inline-formula>h and bias instability (BI) from 0.252°/h to 0.093°/h, with a dynamic range of ±200°/s and scale factor nonlinearity of 259.688 ppm.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26669-26680"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Availability Optimization of a Sensor-Based White Space System in a Data Center Using Markov Process and Metaheuristic Optimization","authors":"Urvashi Godara;Sohan Lal Tyagi;Shikha Bansal","doi":"10.1109/JSEN.2026.3714830","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3714830","url":null,"abstract":"Modern data centers depend on sensor monitoring systems for monitoring environmental and operational metrics, ensuring reliable operation. The failure of sensors and their associated components in the white space monitoring system may substantially impact the system’s overall availability and cause operational failures. This study analyzes the availability behavior of the white space system in the data center to determine how component failures and repair activities affect the system’s overall operational reliability. A stochastic Markov model is developed for the white space system to analyze its performance under different operating and failure conditions. The steady-state availability is obtained from the state transition analysis and is found to be 0.9626. The proposed framework also helps identify subsystem repair priorities and assess the impact of repair facilities on overall system performance. Furthermore, a particle swarm optimization (PSO) algorithm is employed to determine optimal parameter values that enhance system availability. The optimization results show that the system’s availability increased to 0.9999, indicating a significant improvement in overall performance. The proposed approach can improve the availability and operational efficacy of sensor-based monitoring systems in a data center environment.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"25980-25987"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"IEEE Sensors Council","authors":"","doi":"10.1109/JSEN.2026.3724434","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3724434","url":null,"abstract":"","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"C3-C3"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11676271","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871142","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Anatomy-Guided Vision Transformer for Multitask Thermal Gait Analysis in the Timed Up-and-Go Test","authors":"Wei-Lun Chen;Chia-Yeh Hsieh;Kai-Chun Liu;Sheng-Yu Peng;Hung-Yu Wei;Yu Tsao","doi":"10.1109/JSEN.2026.3706259","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3706259","url":null,"abstract":"The Timed Up-and-Go (TUG) test is a widely recognized clinical assessment tool for evaluating lower limb postural stability, dynamic balance, and functional mobility. In recent years, both wearable sensors and vision-based systems have been extensively investigated for TUG assessment. However, conventional sensing modalities, such as inertial sensors and RGB cameras, are limited by issues of signal drift, privacy concerns, or high deployment costs, restricting their applicability in clinical practice. To address these challenges, this study proposes a novel multitask framework based on low-resolution thermal imaging for robust TUG assessment. The core of the framework is the anatomy-guided vision transformer (AG-ViT) encoder, which leverages anatomical priors to adaptively encode keypoint feature maps with enhanced accuracy and computational efficiency. The model performs three tasks, including segmentation of TUG subtasks, recognition of bilateral gait phases, and estimation of temporal gait parameters. Experimental results demonstrate macroaverage <inline-formula> <tex-math>$F1$ </tex-math></inline-formula>-scores of 0.971, 0.948, and 0.941 for TUG subtasks, left-foot gait phases, and right-foot gait phases, respectively. Most temporal gait parameters, including cadence, stance time, swing time, and stride time, exhibited high agreement with ground truth, with Pearson correlation coefficients exceeding 0.8 and intraclass correlation coefficients (ICCs) surpassing 0.75. By combining methodological novelty with privacy-preserving, low-cost sensing, the proposed approach offers strong translational potential for scalable applications in mobility and dynamic balance assessment, including rehabilitation clinics, outpatient assessments, and fall-risk screening programs. To the best of our knowledge, this is the first framework to integrate TUG subtask segmentation, gait phase recognition, and gait parameter estimation using thermal imaging.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26067-26080"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11585904","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871175","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"QF-DWPA: Quantile Feature With Dynamic Windowed Positional Attention for Hard Landing Prediction","authors":"Bokun Chu;Tianqi Chen;Yaotian Jiao;Chang Yu;Linjiang Zheng","doi":"10.1109/JSEN.2026.3715303","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3715303","url":null,"abstract":"Hard landing, as one of the most frequent flight safety incidents during the landing phase, is a major concern for the aviation industry. However, the existing studies mainly focus on one-hot encoded representations to categorize risk levels for hard landing prediction, which fail to capture the slight variations in the maximum vertical acceleration (VRTG) value. To address the above issue, the quantile feature (QF) mechanism and the dynamic windowed positional attention (DWPA) algorithm are incorporated into a vanilla Transformer model to predict specific values of VRTG and capture the peak of VRTG to further predict hard landing and help pilots prevent such exceedance events. First, QF is designed to enhance the model’s ability to grasp the relationship between a single parameter’s value and its quantile level. This is achieved through an exponential-based weighting function that outputs an optimized feature matrix. Moreover, the DWPA algorithm places the most attention on the dynamic window around the predicted landing point across all the training sets and modifies the attention principle of the vanilla Transformer, thereby enhancing VRTG prediction precision during the landing phase. We conducted experiments on real-world quick access recorder (QAR) datasets comprising 572 Airbus A320 flight samples. The experimental results demonstrate that the QF-DWPA Transformer model outperforms other models in hard landing prediction accuracy.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26413-26425"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871196","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Laser Ultrasonic Method for Stress Measurement Based on Grating Laser Generation and Split-Beam Laser Reception","authors":"Shutong Dai;Xiaokai Wang;Kangwen Huang;Yu Peng;Wenlong Yan;Yan Zeng;Rui Zuo;Xiaochun Gu;Xianglin Zhang;Shuang Zhao;Baoming Li","doi":"10.1109/JSEN.2026.3715845","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3715845","url":null,"abstract":"Based on the acoustoelastic effect, traditional ultrasonic residual stress detection technology is susceptible to unstable coupling conditions, resulting in limited measurement accuracy. To address the issue, this article proposes a noncontact laser ultrasonic (LU) residual stress measurement method based on grating laser generation and split-beam laser reception. By integrating a split-beam unit into the laser interferometer architecture, this technique enables simultaneous, synchronized measurement of Rayleigh waves at different locations on the material surface. Under single-pulse laser generation, the system directly acquires one single A-scan waveform containing two Rayleigh wave signals arriving at different times. Subsequently, the autocorrelation algorithm is applied to precisely calculate the time of flight (TOF) from this A-scan waveform, enabling stable measurement of residual stresses. Furthermore, grating laser is employed to generate narrowband Rayleigh waves with minimum dispersion, enabling the mitigation of signal distortion and attenuation, thereby eliminating errors in velocity measurement. Prior to the experiment, a stress–acoustic field coupling model was constructed through COMSOL Multiphysics simulation, validating the feasibility of the ultrasonic stress measurement with narrowband Rayleigh waves using dual-point reception. Rayleigh damping layers were set in the simulation model to eliminate the interference of boundary echoes. In the experiment, an LU testing system with split-beam laser reception and grating generation was established for measuring residual stress. Comparative experiments were designed between split-beam laser reception and single laser reception to verify the stability of split-beam laser reception in TOF measurements when the ultrasonic generation position changes. A U-shaped specimen made of 6061 aluminum alloy, with gradient stress distribution under deformation, served as the test object of stress measurement. To verify the effectiveness of the proposed method in stress measurement, strain gauge measurements were used as a reference. The results indicate good agreement between the results of split-beam LU stress measurement and the strain gauge measurements, validating the potential of the proposed method for precise noncontact residual stress assessment.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"25834-25844"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accurate 3-D Vibration Measurement of Multiple Targets by Using Three Microwave Radars","authors":"Zesheng Ren;Yuyong Xiong;Qiaozhi Song;Xinyu Hui;Zhike Peng","doi":"10.1109/JSEN.2026.3714815","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3714815","url":null,"abstract":"Conventional radar-based vibration sensing is fundamentally limited to 1-D displacement measurement along the sensor’s line-of-sight (LOS), restricting its application in engineering scenarios that require full-field 3-D motion characterization. To this end, this article presents a novel noncontact methodology for multitarget 3-D vibration displacement measurement by leveraging a multistatic network of multiple-input–multiple-output (MIMO) frequency-modulated continuous-wave (FMCW) radars. The proposed system synthesizes the 1-D LOS displacement measurements from three spatially distributed radar units. A dedicated measurement coordinate system is established, and a geometry-based fusion algorithm is derived to reconstruct the complete 3-D displacement vector of each target point from the set of radial measurements. The 3-D reconstruction is formulated as a first-order Taylor expansion of the Jacobian matrix, which is derived from the geometric relationships between the target points and the radar network. The method retains the inherent advantages of microwave sensing, including synchronous multipoint acquisition. Comprehensive experimental validation is conducted. First, a precision validation experiment using a high-precision optical 3-D micrometer as ground truth confirms that the method achieves micrometer-level accuracy in all three spatial dimensions. Then, a multitarget simulation and experiment successfully reconstruct the synchronous 3-D motion of nine target points on a tilting plate, demonstrating the capability for global structural pose tracking. The results verify that the proposed framework effectively overcomes the directional limitation of single-radar systems, providing a reliable and practical solution for advanced applications in structural health monitoring and dynamic mechanical testing where complete 3-D vibrational information is critical.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26361-26371"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spectral Graph Wavelet and Attention-Based Graph Convolutional Network for Bird Sound Recognition","authors":"Xingyuan Wang;Ruichao Yuan;Hang Wang;Qi Yan;Zhongding Fan;Huaiwang Jin;Zhenghao Ke;Yongbin Liu","doi":"10.1109/JSEN.2026.3715990","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3715990","url":null,"abstract":"Bird sound recognition is an effective approach for monitoring the balance of the ecological environment. However, the processing of bird sounds faces the dual challenges of nonstationarity and broad spectral distribution, resulting in difficulties in feature extraction and low recognition accuracy. Hence, in this work, a novel bird sound recognition method is proposed, which is based on spectral graph wavelet attention network (SGWA-Net). First, key frames are selected based on short-time energy and log-mel spectrum, and topological connectivity is established among them to construct a graph-structured representation of bird sound signals. Second, graph convolutional network (GCN) is used to extract high-order global features. Meanwhile, the spectral graph wavelet attention block (SGWA Block) is integrated attention mechanisms with the spectral graph wavelet transform (SGWT) to perform multiscale decomposition, thereby capturing local details across various frequency bands. Finally, the multilayer perceptron (MLP) with a Softmax classifier integrates the multiscale features for recognition of bird species. Experimental results demonstrate that the proposed SGWA-Net model achieves recognition accuracies of 96.23% on a 20-class birdsong dataset and 99.22% on a self-collected 5-class dataset, outperforming conventional methods and offering valuable support for bird vocal diversity monitoring.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26512-26524"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep Learning-Enabled Cross-Device Calibration for Smartphone-Based Optical Biosensing: A Review","authors":"Humeng Zuo;Qingfubo Geng;Hangyu Li;Mengxuan Tao;Yunxuan Feng;Zhaoxin Geng","doi":"10.1109/JSEN.2026.3716330","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3716330","url":null,"abstract":"Smartphone-based optical biosensing has become a compelling route toward portable, low-cost, and accessible analysis, yet its broader translation is still constrained by a central unresolved bottleneck: poor cross-device analytical consistency. Variations in illumination, optics, sensor spectral responsivity, image signal processing (ISP), acquisition geometry, and assay dynamics can transform the same biochemical event into substantially different digital readouts, thereby undermining quantitative reliability. This review recasts deep learning-enabled cross-device calibration as an optical readout harmonization problem for transferable quantification rather than a generic prediction task. We synthesize the field through a unified three-level framework comprising observation calibration to standardize device-dependent images, feature calibration to learn assay-relevant and device-robust representations, and model calibration to impose physics-aware and uncertainty-aware constraints on interpretation. Across smartphone optical biosensing, deep learning shows clear promise, but the evidence base remains uneven, and deployment is still limited less by the absence of models than by weak benchmarking and incomplete validation under device, illumination, lot/time, and field-use shifts. By defining the evidentiary requirements for deployment and providing a practical framework for benchmark design, challenge-oriented evaluation, and revalidation, this review offers a sharper foundation for developing reliable, scalable, and analytically defensible smartphone biosensing systems.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"25373-25391"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871366","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}