Sensors

11 pages, 1902 KiB

Open AccessCommunication

Innovative Pedicle Screw Insertion with Mixed Reality Technology Improves Insertion Accuracy in Spinal Surgery

by Shintaro Obata, Akira Shinohara, Daigo Arimura, Shunsuke Katsumi, Hiroki Wakiya and Mitsuru Saito

Sensors 2025, 25(13), 3939; https://doi.org/10.3390/s25133939 - 24 Jun 2025

Abstract

The accuracy of pedicle screw insertion in pediatric scoliosis correction surgery using augmented reality technology in combination with a conventional navigation system was evaluated, and its usefulness was verified. A retrospective study of patients who underwent mixed reality technology-assisted posterior scoliosis correction and [...] Read more.

The accuracy of pedicle screw insertion in pediatric scoliosis correction surgery using augmented reality technology in combination with a conventional navigation system was evaluated, and its usefulness was verified. A retrospective study of patients who underwent mixed reality technology-assisted posterior scoliosis correction and fixation was conducted. In total, 361 pedicle screws inserted with a mixed reality technology-assisted navigation system were analyzed; 25 pedicle screws (6.9%) showed Rao Classification Grade 1 deviation, whereas 0.83% showed Rao Classification Grade 2.3 deviation, which is a clinical deviation. In terms of the relationship between the rotation of the vertebral body and the deviation of the pedicle screw, the pedicle screw tended to deviate more easily when it was necessary to insert the pedicle screw in a more strongly oblique position due to the rotation of the vertebral body. The results suggest that the pedicle screw insertion accuracy with augmented reality technology may be superior to that with conventional navigation alone in scoliosis correction and fusion surgery for scoliosis in children. This system is expected to become a standard support tool for spine surgery and will contribute to improving the success rate of surgery and reducing the burden on the surgeon. Full article

(This article belongs to the Section Biomedical Sensors)

► Show Figures

Figure 1

18 pages, 6279 KiB

Open AccessArticle

Fabrication and Characterization of a PZT-Based Touch Sensor Using Combined Spin-Coating and Sputtering Methods

by Melih Ozden, Omer Coban and Tevhit Karacali

Sensors 2025, 25(13), 3938; https://doi.org/10.3390/s25133938 - 24 Jun 2025

Abstract

This study presents the successful fabrication of lead zirconate titanate (PZT) thin films on silicon (Si) substrates using a hybrid deposition method combining spin-coating and RF sputtering techniques. Initially, a PZT layer was deposited through four successive spin-coating cycles, followed by an additional [...] Read more.

This study presents the successful fabrication of lead zirconate titanate (PZT) thin films on silicon (Si) substrates using a hybrid deposition method combining spin-coating and RF sputtering techniques. Initially, a PZT layer was deposited through four successive spin-coating cycles, followed by an additional layer formed via RF sputtering. The resulting multilayer structure was annealed at 700

^{\circ}

C for 2 h to improve crystallinity. Comprehensive material characterization was conducted using XRD, SEM, cross-sectional SEM, EDX, and UV–VIS absorbance spectroscopy. The analyses confirmed the formation of a well-crystallized perovskite phase, a uniform surface morphology, and an optical band gap of approximately 3.55 eV, supporting its suitability for sensing applications. Building upon these findings, a multilayer PZT-based touch sensor was fabricated and electrically characterized. Low-frequency I–V measurements demonstrated consistent and repeatable polarization behavior under cyclic loading conditions. In addition,

| Z |

–f measurements were performed to assess the sensor’s dynamic electrical behavior. Although expected dielectric responses were observed, the absence of distinct anti-resonance peaks suggested non-idealities linked to Ag

^{+}

ion diffusion from the electrode layers. To account for these effects, the classical Butterworth–Van Dyke (BVD) equivalent circuit model was extended with additional inductive and resistive components representing parasitic pathways. This modified model provided excellent agreement with the measured impedance and phase data, offering deeper insight into the interplay between material degradation and electrical performance. Overall, the developed sensor structure exhibits strong potential for use in piezoelectric sensing applications, particularly for tactile and pressure-based interfaces. Full article

(This article belongs to the Section Sensor Materials)

17 pages, 1448 KiB

Open AccessArticle

A Data Reconstruction Method for Inspection Mode in GBSAR Monitoring Using Sage–Husa Adaptive Kalman Filtering and RTS Smoothing

by Yaolong Qi, Jialiang Guo, Jiaxin Hui, Ting Hou, Pingping Huang, Weixian Tan and Wei Xu

Sensors 2025, 25(13), 3937; https://doi.org/10.3390/s25133937 - 24 Jun 2025

Abstract

Ground-based synthetic aperture radar (GBSAR) has been widely used in the fields of early warning of geologic hazards and deformation monitoring of engineering structures due to its characteristics of high spatial resolution, zero spatial baseline, and short revisit period. However, in the continuous [...] Read more.

Ground-based synthetic aperture radar (GBSAR) has been widely used in the fields of early warning of geologic hazards and deformation monitoring of engineering structures due to its characteristics of high spatial resolution, zero spatial baseline, and short revisit period. However, in the continuous monitoring process of GBSAR, due to the sudden failure of radar equipment, such as power failure, or the influence of alternating work between multiple regions, it often leads to discontinuous data collection, and this problem caused by missing data is collectively called “inspection mode”. The problem of missing data in the inspection mode not only destroys the spatial and temporal continuity of the data but also affects the accuracy of the subsequent deformation analysis. In order to solve this problem, in this paper, we propose a data reconstruction method that combines Sage–Husa Kalman adaptive filtering and the Rauch–Tung–Striebel (RTS) smoothing algorithm. The method is based on the principle of Kalman filtering and solves the problem of “model mismatch” caused by the fixed noise statistics of traditional Kalman filtering by dynamically adjusting the noise covariance to adapt to the non-stationary characteristics of the observed data. Subsequently, the Rauch–Tung–Striebel (RTS) smoothing algorithm is used to process the preliminary filtering results to eliminate the cumulative error during the period of missing data and recover the complete and smooth deformation time series. The experimental and simulation results show that this method successfully restores the spatial and temporal continuity of the inspection data, thus improving the overall accuracy and stability of deformation monitoring. Full article

(This article belongs to the Section Remote Sensors)

30 pages, 2180 KiB

Open AccessArticle

WiPIHT: A WiFi-Based Position-Independent Passive Indoor Human Tracking System

by Xu Xu, Xilong Che, Xianqiu Meng, Long Li, Ziqi Liu and Shuai Shao

Sensors 2025, 25(13), 3936; https://doi.org/10.3390/s25133936 - 24 Jun 2025

Abstract

Unlike traditional vision-based camera tracking, human indoor localization and activity trajectory recognition also employ other methods such as infrared tracking, acoustic localization, and locators. These methods have significant environmental limitations or dependency on specialized equipment. Currently, WiFi-based human sensing is a novel and [...] Read more.

Unlike traditional vision-based camera tracking, human indoor localization and activity trajectory recognition also employ other methods such as infrared tracking, acoustic localization, and locators. These methods have significant environmental limitations or dependency on specialized equipment. Currently, WiFi-based human sensing is a novel and important method for human activity recognition. However, most WiFi-based activity recognition methods have limitations, such as using WiFi fingerprints to identify human activities. They either require extensive sample collection and training, are constrained by a fixed environmental layout, or rely on the precise positioning of transmitters (TXs) and receivers (RXs) within the space. If the positions are uncertain, or change, the sensing performance becomes unstable. To address the dependency of current WiFi indoor human activity trajectory reconstruction on the TX-RX position, we propose WiPIHT, a stable system for tracking indoor human activity trajectories using a small number of commercial WiFi devices. This system does not require additional hardware to be carried or locators to be attached, enabling passive, real-time, and accurate tracking and trajectory reconstruction of indoor human activities. WiPIHT is based on an innovative CSI channel analysis method, analyzing its autocorrelation function to extract location-independent real-time movement speed features of the human body. It also incorporates Fresnel zone and motion velocity direction decomposition to extract movement direction change patterns independent of the relative position between the TX-RX and the human body. By combining real-time speed and direction curve features, the system derives the shape of the human movement trajectory. Experiments demonstrate that, compared to existing methods, our system can accurately reconstruct activity trajectory shapes even without knowing the initial positions of the TX or the human body. Additionally, our system shows significant advantages in tracking accuracy, real-time performance, equipment, and cost. Full article

(This article belongs to the Special Issue Recent Advances in Smart Mobile Sensing Technology)

28 pages, 2284 KiB

Open AccessArticle

Validation of a Low-Cost Open-Ended Coaxial Probe Setup for Broadband Permittivity Measurements up to 6 GHz

by Julia Arias-Rodríguez, Raúl Moreno-Merín, Andrea Martínez-Lozano, Germán Torregrosa-Penalva and Ernesto Ávila-Navarro

Sensors 2025, 25(13), 3935; https://doi.org/10.3390/s25133935 - 24 Jun 2025

Abstract

This work presents the validation of a low-cost measurement system based on an open-ended coaxial SMA (SubMiniature version A) probe for the characterization of complex permittivity in the microwave frequency range. The system combines a custom-fabricated probe, a vector network analyzer, and a [...] Read more.

This work presents the validation of a low-cost measurement system based on an open-ended coaxial SMA (SubMiniature version A) probe for the characterization of complex permittivity in the microwave frequency range. The system combines a custom-fabricated probe, a vector network analyzer, and a dedicated software application that implements three analytical models: capacitive, radiation, and virtual transmission line models. A comprehensive experimental campaign was carried out involving pure polar liquids, saline solutions, and biological tissues, with the measurements compared against those obtained using a high-precision commercial probe. The results confirm that the proposed system is capable of delivering accurate and reproducible permittivity values up to at least 6 GHz. Among the implemented models, the radiation model demonstrated the best overall performance, particularly in biological samples. Additionally, reproducibility tests with three independently assembled SMA probes showed normalized deviations below 3%, confirming the robustness of the design. These results demonstrate that the proposed system constitutes a viable alternative for cost-sensitive applications requiring portable or scalable microwave dielectric characterization. Full article

(This article belongs to the Special Issue Advanced Microwave Sensors and Their Applications in Measurement)

20 pages, 773 KiB

Open AccessArticle

Relay Selection for Covert Communication with an Active Warden

by Jong Yeol Ryu and Jung Hoon Lee

Sensors 2025, 25(13), 3934; https://doi.org/10.3390/s25133934 - 24 Jun 2025

Abstract

In this paper, we consider covert communication with multiple relays and an active warden who not only sends jamming signals but also aims to detect the covert transmission. In the relay system with the active warden, the most critical factor is the channel [...] Read more.

In this paper, we consider covert communication with multiple relays and an active warden who not only sends jamming signals but also aims to detect the covert transmission. In the relay system with the active warden, the most critical factor is the channel between the relay and the warden, as the warden leverages this channel to transmit jamming signals while trying to detect the presence of covert communication. To mitigate the impact of the active warden, we propose a relay selection scheme that selects the relay with the minimum channel gain to the warden. We analyze the performance of the proposed scheme and demonstrate how increasing the number of relays leads to performance improvements based on analytical results. Numerical results show that the analytical predictions closely match the simulations, and our proposed scheme effectively increases the covert rate while minimizing the threat posed by the active warden. Full article

(This article belongs to the Special Issue Security and Privacy in Next-Generation Mobile Networks: Challenges, Solutions, and Innovations)

27 pages, 2484 KiB

Open AccessArticle

Hierarchical Torque Vectoring Control Strategy of Distributed Driving Electric Vehicles Considering Stability and Economy

by Shuiku Liu, Haichuan Zhang, Shu Wang and Xuan Zhao

Sensors 2025, 25(13), 3933; https://doi.org/10.3390/s25133933 - 24 Jun 2025

Abstract

Coordinating vehicle handling stability and energy consumption remains a key challenge for distributed driving electric vehicles (DDEVs). In this paper, a hierarchical torque vectoring control strategy is proposed to address this issue. First, a tire road friction coefficient (TRFC) estimator based on the [...] Read more.

Coordinating vehicle handling stability and energy consumption remains a key challenge for distributed driving electric vehicles (DDEVs). In this paper, a hierarchical torque vectoring control strategy is proposed to address this issue. First, a tire road friction coefficient (TRFC) estimator based on the fusion of vision and dynamic is developed to accurately and promptly obtain the TRFC in the upper layer. Second, a direct yaw moment control (DYC) strategy based on the adaptive model predictive control (MPC) is designed to ensure vehicle stability in the middle layer, where tire cornering stiffness is updated dynamically based on the estimated TRFC. Then, the lower layer develops the torque vectoring allocation controller, which comprehensively considers handling stability and energy consumption and distributes the driving torques among the wheels. The weight between stability and economy is coordinated according to the stability boundaries derived from an extended phase-plane correlated with the TRFC. Finally, Hardware-in-the-Loop (HIL) simulations are conducted to validate the effectiveness of the proposed strategy. The results demonstrate that compared with the conventional stability torque distribution strategy, the proposed control strategy not only reduces the RMSE of sideslip angle by 44.88% but also decreases the motor power consumption by 24.45% under DLC conditions, which indicates that the proposed method can significantly enhance vehicle handling stability while reducing energy consumption. Full article

(This article belongs to the Section Vehicular Sensing)

16 pages, 499 KiB

Open AccessArticle

Adaptive Sampling Framework for Imbalanced DDoS Traffic Classification

by Hongjoong Kim, Deokhyeon Ham and Kyoung-Sook Moon

Sensors 2025, 25(13), 3932; https://doi.org/10.3390/s25133932 - 24 Jun 2025

Abstract

Imbalanced data is a major challenge in network security applications, particularly in DDoS (Distributed Denial of Service) traffic classification, where detecting minority classes is critical for timely and cost-effective defense. Existing machine learning and deep learning models often fail to accurately classify such [...] Read more.

Imbalanced data is a major challenge in network security applications, particularly in DDoS (Distributed Denial of Service) traffic classification, where detecting minority classes is critical for timely and cost-effective defense. Existing machine learning and deep learning models often fail to accurately classify such underrepresented attack types, leading to significant degradation in performance. In this study, we propose an adaptive sampling strategy that combines oversampling and undersampling techniques to address the class imbalance problem at the data level. We evaluated our approach using benchmark DDoS traffic datasets, where it demonstrated improved classification performance across key metrics, including accuracy, recall, and F1-score, compared to baseline models and conventional sampling methods. The results indicate that the proposed adaptive sampling approach improved minority class detection performance under the tested conditions, thereby improving the reliability of sensor-driven security systems. This work contributes a robust and adaptable method for imbalanced data classification, with potential applications across simulated sensor environments where anomaly detection is essential. Full article

(This article belongs to the Special Issue Feature Papers in Fault Diagnosis & Sensors 2025)

► Show Figures

Figure 1

14 pages, 1800 KiB

Open AccessArticle

Design of a Photonic Crystal Fiber Optic Magnetic Field Sensor Based on Surface Plasmon Resonance

by Yuxuan Yi, Hua Yang, Tangyou Sun, Zao Yi, Zigang Zhou, Chao Liu and Yougen Yi

Sensors 2025, 25(13), 3931; https://doi.org/10.3390/s25133931 - 24 Jun 2025

Abstract

To enhance the sensing performance of fiber-optic magnetic field sensors, we explored the design, optimization, and application prospects of a D-type fiber-optic magnetic field sensor. This D-type PCF-SPR sensor is metal coated on one side (the metal used in this study is gold), [...] Read more.

To enhance the sensing performance of fiber-optic magnetic field sensors, we explored the design, optimization, and application prospects of a D-type fiber-optic magnetic field sensor. This D-type PCF-SPR sensor is metal coated on one side (the metal used in this study is gold), which serves as the active metal for SPR and enhances structural stability. Magnetic fluid is applied on the outer side of the gold film for SPR magnetic field sensing. Six internal air holes arranged in a hexagonal shape form a central light transmission channel that facilitates the connection between the two modes, which are the sensor’s core mode and SPP mode, respectively. The outer six large air holes and two small air holes are arranged in a circular pattern to form the cladding, which allows for better energy transmission and reduces energy loss in the fiber. In this paper, the finite element method is employed to analyze the transmission performance of the sensor, focusing on the transmission mode. Guidelines for optimizing the PCF-SPR sensor are derived from analyzing the fiber optic sensor’s dispersion curve, the impact of surface plasmon excitation mode, and the core mode energy on sensing performance. After analyzing and optimizing the transmission mode and structural parameters, the optimized sensor achieves a magnetic field sensitivity of 18,500 pm/mT and a resolution of 54 nT. This performance is several orders of magnitude higher than most other sensors in terms of sensitivity and resolution. The SPR-PCF magnetic field sensor offers highly sensitive and accurate magnetic field measurements and shows promising applications in medical and industrial fields. Full article

(This article belongs to the Special Issue Advances and Applications of Magnetic Sensors: 2nd Edition)

► Show Figures

Figure 1

16 pages, 1166 KiB

Open AccessEssay

Research on Acoustic Scene Classification Based on Time–Frequency–Wavelet Fusion Network

by Fengzheng Bi and Lidong Yang

Sensors 2025, 25(13), 3930; https://doi.org/10.3390/s25133930 - 24 Jun 2025

Abstract

Acoustic scene classification aims to recognize the scenes corresponding to sound signals in the environment, but audio differences from different cities and devices can affect the model’s accuracy. In this paper, a time–frequency–wavelet fusion network is proposed to improve model performance by focusing [...] Read more.

Acoustic scene classification aims to recognize the scenes corresponding to sound signals in the environment, but audio differences from different cities and devices can affect the model’s accuracy. In this paper, a time–frequency–wavelet fusion network is proposed to improve model performance by focusing on three dimensions: the time dimension of the spectrogram, the frequency dimension, and the high- and low-frequency information extracted by a wavelet transform through a time–frequency–wavelet module. Multidimensional information was fused through the gated temporal–spatial attention unit, and the visual state space module was introduced to enhance the contextual modeling capability of audio sequences. In addition, Kolmogorov–Arnold network layers were used in place of multilayer perceptrons in the classifier part. The experimental results show that the proposed method achieves a 56.16% average accuracy on the TAU Urban Acoustic Scenes 2022 mobile development dataset, which is an improvement of 6.53% compared to the official baseline system. This performance improvement demonstrates the effectiveness of the model in complex scenarios. In addition, the accuracy of the proposed method on the UrbanSound8K dataset reached 97.60%, which is significantly better than the existing methods, further verifying the generalization ability of the proposed model in the acoustic scene classification task. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

18 pages, 18889 KiB

Open AccessArticle

A Handheld Multispectral Device for Assessing Leaf Nitrogen Concentrations in Maize

by Felipe Hermínio Meireles Nogueira, Adunias dos Santos Teixeira, Sharon Gomes Ribeiro, Luís Clênio Jario Moreira, Odílio Coimbra da Rocha Neto, Fernando Bezerra Lopes and Ricardo Emílio Ferreira Quevedo Nogueira

Sensors 2025, 25(13), 3929; https://doi.org/10.3390/s25133929 - 24 Jun 2025

Abstract

This study presents the MSPAT (Multispectral Soil Plant Analysis Tool), a device designed for assessing leaf nitrogen concentrations in maize crops under field conditions. The MSPAT includes the AS7265x sensor, which has 18 bands and covers the spectrum from 410 to 940 nm. [...] Read more.

This study presents the MSPAT (Multispectral Soil Plant Analysis Tool), a device designed for assessing leaf nitrogen concentrations in maize crops under field conditions. The MSPAT includes the AS7265x sensor, which has 18 bands and covers the spectrum from 410 to 940 nm. This device was designed to be portable, using the ESP32 microcontroller and incorporating such functionalities as data storage on a MicroSD card, communication with a smartphone via Wi-Fi, and geolocation of acquired data. The MSPAT was evaluated in an experiment conducted at the Federal University of Ceará (UFC), where maize was subjected to different doses of nitrogen fertiliser (0, 60, 90, 120, 150, and 180 kg·ha⁻¹ N). Spectral readings were taken at three phenological stages (V5, V10, and R2) using the MSPAT, an SPAD-502 chlorophyll meter, and a FieldSpec PRO FR3 spectroradiometer. After the optical measurements were taken, the nitrogen concentrations in the leaves were determined in a laboratory by using the Kjeldahl method. The data analysis included the calculation of normalised ratio indices (NRIs) using linear regression and the application of multivariate statistical methods (PLSR and PCR) for predicting leaf nitrogen concentrations (LNCs). The best performance for the MSPAT index (NRI) was obtained using the 900 nm and the 560 nm bands (R² = 0.64) at stage V10. In the validation analysis, the MSPAT presented an R² of 0.79, showing performance superior to that of SPAD-502, which achieved an R² of 0.70. This confirms the greater potential of the MSPAT compared to commercial equipment and makes it possible to obtain results similar to those obtained using the reference spectroradiometer. The PLSR model with data from the FieldSpec 3 provided important validation metrics when using reflectance data with first-derivative transformation (R² = 0.88, RMSE = 1.94 and MAE = 1.28). When using the MSPAT, PLSR (R² = 0.75, RMSE = 2.77 and MAE = 2.26) exhibited values of metrics similar to those for PCR (R² = 0.75, RMSE = 2.78 and MAE = 2.26). This study validates the use of MSPAT as an effective tool for monitoring the nutritional status of maize to optimize the use of nitrogen fertilisers. Full article

(This article belongs to the Special Issue Hyperspectral Sensing: Imaging and Applications)

► Show Figures

Figure 1

19 pages, 2410 KiB

Open AccessArticle

MAK-Net: A Multi-Scale Attentive Kolmogorov–Arnold Network with BiGRU for Imbalanced ECG Arrhythmia Classification

by Cong Zhao, Bingwei Lai, Yongzheng Xu, Yiping Wang and Haorong Dong

Sensors 2025, 25(13), 3928; https://doi.org/10.3390/s25133928 - 24 Jun 2025

Abstract

Accurate classification of electrocardiogram (ECG) signals is vital for reliable arrhythmia diagnosis and informed clinical decision-making, yet real-world datasets often suffer severe class imbalance that degrades recall and F1-score. To address these limitations, we introduce MAK-Net, a hybrid deep learning framework that combines: [...] Read more.

Accurate classification of electrocardiogram (ECG) signals is vital for reliable arrhythmia diagnosis and informed clinical decision-making, yet real-world datasets often suffer severe class imbalance that degrades recall and F1-score. To address these limitations, we introduce MAK-Net, a hybrid deep learning framework that combines: (1) a four-branch multiscale convolutional module for comprehensive feature extraction across diverse waveform morphologies; (2) an efficient channel attention mechanism for adaptive weighting of clinically salient segments; (3) bidirectional gated recurrent units (BiGRU) to capture long-range temporal dependencies; and (4) Kolmogorov–Arnold Network (KAN) layers with learnable spline activations for enhanced nonlinear representation and interpretability. We further mitigate imbalance by synergistically applying focal loss and the Synthetic Minority Oversampling Technique (SMOTE). On the MIT-BIH arrhythmia database, MAK-Net attains state-of-the-art performance—0.9980 accuracy, 0.9888 F1-score, 0.9871 recall, 0.9905 precision, and 0.9991 specificity—demonstrating superior robustness to imbalanced classes compared with existing methods. These findings validate the efficacy of multiscale feature fusion, attention-guided learning, and KAN-based nonlinear mapping for automated, clinically reliable arrhythmia detection. Full article

(This article belongs to the Section Biomedical Sensors)

► Show Figures

Figure 1

34 pages, 3133 KiB

Open AccessReview

Review of State Estimation Methods for Autonomous Ground Vehicles: Perspectives on Estimation Objects, Vehicle Characteristics, and Key Algorithms

by Xiaoyu Wang, Te Chen, Renzhong Wang, Jiankang Lu and Guowei Dou

Sensors 2025, 25(13), 3927; https://doi.org/10.3390/s25133927 - 24 Jun 2025

Abstract

This paper reviews research on vehicle driving state estimation research. Based on the discussion of the importance, development history, and application fields of this topic of research, it focuses on analyzing vehicle state estimation techniques from different perspectives, namely (1) from the perspective [...] Read more.

This paper reviews research on vehicle driving state estimation research. Based on the discussion of the importance, development history, and application fields of this topic of research, it focuses on analyzing vehicle state estimation techniques from different perspectives, namely (1) from the perspective of the estimation objects, including vehicle attitude and driving state estimations, chassis component key dynamic parameter estimations, and vehicle driving environment state estimations; (2) from the perspective of vehicle characteristics, including vehicle dynamics coupling characteristics, vehicle multi-source information redundancy characteristics, and vehicle state transition characteristics; (3) from the perspective of key estimation algorithms, including model-based Kalman filtering algorithms, data-driven machine learning algorithms, and optimization estimation algorithms combining mechanism-based and data-driven approaches. This manuscript helps interested readers to comprehensively understand the research progress, technical features, and future trends of vehicle state estimation technology from the perspective of overall architecture and subdomains. Full article

(This article belongs to the Special Issue Sensor Fusion and Advanced Controller for Connected and Automated Vehicles (Volume II))

► Show Figures

Figure 1

23 pages, 4696 KiB

Open AccessArticle

A Hybrid Compact Convolutional Transformer with Bilateral Filtering for Coffee Berry Disease Classification

by Biniyam Mulugeta Abuhayi and Andras Hajdu

Sensors 2025, 25(13), 3926; https://doi.org/10.3390/s25133926 - 24 Jun 2025

Abstract

Coffee berry disease (CBD), caused by Colletotrichum kahawae, significantly threatens global Coffee arabica production, leading to major yield losses. Traditional detection methods are often subjective and inefficient, particularly in resource-limited settings. While deep learning has advanced plant disease detection, most existing research targets [...] Read more.

Coffee berry disease (CBD), caused by Colletotrichum kahawae, significantly threatens global Coffee arabica production, leading to major yield losses. Traditional detection methods are often subjective and inefficient, particularly in resource-limited settings. While deep learning has advanced plant disease detection, most existing research targets leaf diseases, with limited focus on berry-specific infections like CBD. This study proposes a lightweight and accurate solution using a Compact Convolutional Transformer (CCT) for classifying healthy and CBD-affected coffee berries. The CCT model combines parallel convolutional branches for hierarchical feature extraction with a transformer encoder to capture long-range dependencies, enabling high performance on limited data. A dataset of 1737 coffee berry images was enhanced using bilateral filtering and color segmentation. The CCT model, integrated with a Multilayer Perceptron (MLP) classifier and optimized through early stopping and regularization, achieved a validation accuracy of 97.70% and a sensitivity of 100% for CBD detection. Additionally, CCT-extracted features performed well with traditional classifiers, including Support Vector Machine (SVM) (82.47% accuracy; AUC 0.91) and Decision Tree (82.76% accuracy; AUC 0.86). Compared to pretrained models, the proposed system delivered superior accuracy (97.5%) with only 0.408 million parameters and faster training (2.3 s/epoch), highlighting its potential for real-time, low-resource deployment in sustainable coffee production systems. Full article

(This article belongs to the Special Issue Feature Papers in Smart Agriculture 2025)

► Show Figures

Figure 1

23 pages, 5579 KiB

Open AccessArticle

End-to-End Interrupted Sampling Repeater Jamming Countermeasure Network Under Low Signal-to-Noise Ratio

by Gane Dai, Xiaoxuan Yang, Sha Huan, Ziyang Chen, Cong Peng and Yuanqin Xu

Sensors 2025, 25(13), 3925; https://doi.org/10.3390/s25133925 - 24 Jun 2025

Abstract

Interrupted sampling repeater jamming (ISRJ) is characterized by its coherent processing gains and flexible modulation techniques. ISRJ generates spurious targets along the range, which presents significant challenges to the radar systems. However, existing ISRJ countermeasure methods struggle to eliminate ISRJ signals without compromising [...] Read more.

Interrupted sampling repeater jamming (ISRJ) is characterized by its coherent processing gains and flexible modulation techniques. ISRJ generates spurious targets along the range, which presents significant challenges to the radar systems. However, existing ISRJ countermeasure methods struggle to eliminate ISRJ signals without compromising the integrity of the real target signal, especially under low-signal-to-noise-ratio (SNR) conditions, resulting in a deteriorated sidelobe and diminished detection performance. We propose a complex-valued encoder–decoder network (CVEDNet) to address these challenges based on signal decomposition. This network offers an end-to-end ISRJ suppression approach, working on complex-valued time-domain signals without the need for additional preprocessing. The encoding and decoding structure suppresses noise components and obtains more compact echo feature representations through layer-by-layer compression and reconstruction. A stacked dual-branch structure and multi-scale dilated convolutions are adopted to further separate the echo signal and ISRJ based on high-dimensional features. A multi-domain combined loss function integrates the waveform and range-pulse-compression information to ensure the amplitude and phase integrity of the reconstructed echo waveform during the training process. The effectiveness of the proposed method was validated in terms of its jamming suppression capability, echo fidelity, and detection performance indicators under low-SNR conditions compared to conventional methods. Full article

(This article belongs to the Special Issue Detection, Recognition and Identification in the Radar Applications)

► Show Figures

Figure 1

16 pages, 2246 KiB

Open AccessArticle

Context-Aware Beam Selection for IRS-Assisted mmWave V2I Communications

by Ricardo Suarez del Valle, Abdulkadir Kose and Haeyoung Lee

Sensors 2025, 25(13), 3924; https://doi.org/10.3390/s25133924 - 24 Jun 2025

Abstract

Millimeter wave (mmWave) technology, with its ultra-high bandwidth and low latency, holds significant promise for vehicle-to-everything (V2X) communications. However, it faces challenges such as high propagation losses and limited coverage in dense urban vehicular environments. Intelligent Reflecting Surfaces (IRSs) help address these issues [...] Read more.

Millimeter wave (mmWave) technology, with its ultra-high bandwidth and low latency, holds significant promise for vehicle-to-everything (V2X) communications. However, it faces challenges such as high propagation losses and limited coverage in dense urban vehicular environments. Intelligent Reflecting Surfaces (IRSs) help address these issues by enhancing mmWave signal paths around obstacles, thereby maintaining reliable communication. This paper introduces a novel Contextual Multi-Armed Bandit (C-MAB) algorithm designed to dynamically adapt beam and IRS selections based on real-time environmental context. Simulation results demonstrate that the proposed C-MAB approach significantly improves link stability, doubling average beam sojourn times compared to traditional SNR-based strategies and standard MAB methods, and achieving gains of up to four times the performance in scenarios with IRS assistance. This approach enables optimized resource allocation and significantly improves coverage, data rate, and resource utilization compared to conventional methods. Full article

(This article belongs to the Topic Machine Learning in Communication Systems and Networks, 2nd Edition)

► Show Figures

Figure 1

21 pages, 1763 KiB

Open AccessArticle

Machine Learning-Based Predictive Maintenance at Smart Ports Using IoT Sensor Data

by Sheraz Aslam, Alejandro Navarro, Andreas Aristotelous, Eduardo Garro Crevillen, Alvaro Martınez-Romero, Álvaro Martínez-Ceballos, Alessandro Cassera, Kyriacos Orphanides, Herodotos Herodotou and Michalis P. Michaelides

Sensors 2025, 25(13), 3923; https://doi.org/10.3390/s25133923 - 24 Jun 2025

Abstract

Maritime transportation plays a critical role in global containerized cargo logistics, with seaports serving as key nodes in this system. Ports are responsible for container loading and unloading, along with inspection, storage, and timely delivery to the destination, all of which heavily depend [...] Read more.

Maritime transportation plays a critical role in global containerized cargo logistics, with seaports serving as key nodes in this system. Ports are responsible for container loading and unloading, along with inspection, storage, and timely delivery to the destination, all of which heavily depend on the performance of the container handling equipment (CHE). Inefficient maintenance strategies and unplanned maintenance of the port equipment can lead to operational disruptions, including unexpected delays and long waiting times in the supply chain. Therefore, the maritime industry must adopt intelligent maintenance strategies at the port to optimize operational efficiency and resource utilization. Towards this end, this study presents a machine learning (ML)-based approach for predicting faults in CHE to improve equipment reliability and overall port performance. Firstly, a statistical model was developed to check the status and health of the hydraulic system, as it is crucial for the operation of the machines. Then, several ML models were developed, including artificial neural networks (ANNs), decision trees (DTs), random forest (RF), Extreme Gradient Boosting (XGBoost), and Gaussian Naive Bayes (GNB) to predict inverter over-temperature faults due to fan failures, clogged filters, and other related issues. From the tested models, the ANNs achieved the highest performance in predicting the specific faults with a 98.7% accuracy and 98.0% F1-score. Full article

(This article belongs to the Special Issue Sensors and IoT Technologies for the Smart Industry)

► Show Figures

Figure 1

12 pages, 1876 KiB

Open AccessArticle

High-Precision and Low-Complexity Silicon Waveguide-Integrated Temperature Sensor System

by Zhiming Zhang, Haole Kong and Yi Li

Sensors 2025, 25(13), 3922; https://doi.org/10.3390/s25133922 - 24 Jun 2025

Abstract

This paper proposes a high-sensitivity-integrated temperature sensor with low complexity based on a silicon waveguide. The waveguide layout is optimized through the finite-difference time-domain (FDTD) simulations, and a compressed taper structure improves the efficiency of speckle data collection while reducing the system complexity [...] Read more.

This paper proposes a high-sensitivity-integrated temperature sensor with low complexity based on a silicon waveguide. The waveguide layout is optimized through the finite-difference time-domain (FDTD) simulations, and a compressed taper structure improves the efficiency of speckle data collection while reducing the system complexity and cost. To achieve precise temperature demodulation, this paper employed a convolutional neural network (CNN) for nonlinear fitting. Experimental results demonstrate the sensor’s ability to perform temperature measurement in the range of −20 °C to 100 °C, with a best resolution of 0.00287 °C (2.87 mK). The resolution and reliability of the measurements are validated by comparison with the theoretical values. This study introduces a novel approach to silicon waveguide-based temperature sensing. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Figure 1

12 pages, 7858 KiB

Open AccessArticle

Strain Monitoring of Vertical Axis Wind Turbine Tower Using Fiber Bragg Gratings

by Bastien Van Esbeen, Valentin Manto, Damien Kinet, Corentin Guyot and Christophe Caucheteur

Sensors 2025, 25(13), 3921; https://doi.org/10.3390/s25133921 - 24 Jun 2025

Abstract

This article presents the findings of an experimental study conducted on a vertical axis wind turbine (VAWT) tower instrumented with cascaded fiber Bragg grating (FBG) sensors to detect bending deformations. Structural health monitoring (SHM) is an essential need in the industry to reduce [...] Read more.

This article presents the findings of an experimental study conducted on a vertical axis wind turbine (VAWT) tower instrumented with cascaded fiber Bragg grating (FBG) sensors to detect bending deformations. Structural health monitoring (SHM) is an essential need in the industry to reduce costs and maintenance time, and to prevent machine failures. First, FBG strain sensors were glued vertically along the tower to investigate the sensors behavior as a function of their height. The maximum signal-to-noise ratio is obtained when FBGs are placed at the tower base. Then, four packages were installed inside the tower, at the base, according to four cardinal directions. Each package contains an FBG strain sensor, and an extra temperature FBG for discrimination. The use of easy-to-deploy packages is a must for industrial installations. Afterwards, by using power spectral density (PSD) on the strain signals, three sources of tower oscillations are discovered: wind force, structure unbalance, and 1st tower mode resonance, each with its intrinsic frequency. Wind force and structure unbalance cause mechanical stresses at a frequency proportional to the wind turbine rotational speed, while the 1st tower mode frequency depends only on the machine geometry, regardless of the rotational speed. This study also analyzes the deformation amplitude for different rotational rates within the VAWT operational range (10–35 rpm). The resonance amplitude depends on the proximity of the rotational rate to the resonant frequency (22 rpm) and the duration at that rate. For structure unbalance, the oscillation amplitude increases with the rotational rate, due to the centrifugal effect. It is supposed that wind force deformation amplitude naturally depends on wind speed, which is unpredictable at a given precise time. The results of our experimental observations are very valuable for both the wind turbine manufacturer and owner. Full article

(This article belongs to the Section Physical 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