Pub Date : 2025-02-27DOI: 10.1109/JSEN.2025.3543343
Yuqiang He;Jun Wang;Yaxin Li;Yuquan Luo
In this article, we introduce a novel method for human skeletal joint localization using millimeter-wave (mmWave) radar, effectively overcoming the limitations of vision-based pose estimation methods, which are vulnerable to changes in lighting conditions and pose privacy concerns. The method leverages mmWave radar to generate 4-D time-series point cloud data, which is then projected onto the depth-azimuth and depth-elevation planes. This projection helps mitigate the sparsity inherent in traditional point cloud data and reduces the complexity of the machine learning model required for pose estimation. The input data structure is optimized using a sliding window technique, where consecutive frames are processed by a convolutional neural network (CNN) to extract spatial features. These features are then sorted chronologically and fed into a bi-directional long short-term memory (BiLSTM) to capture temporal features, resulting in the accurate localization of 25 skeletal joints. To validate the performance and effectiveness of the proposed method, we created a dataset comprising three body types and ten distinct actions. The experimental results demonstrate the method’s outstanding human pose estimation capability.
{"title":"Dual-Path CNN–BiLSTM for mmWave-Based Human Skeletal Pose Estimation","authors":"Yuqiang He;Jun Wang;Yaxin Li;Yuquan Luo","doi":"10.1109/JSEN.2025.3543343","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3543343","url":null,"abstract":"In this article, we introduce a novel method for human skeletal joint localization using millimeter-wave (mmWave) radar, effectively overcoming the limitations of vision-based pose estimation methods, which are vulnerable to changes in lighting conditions and pose privacy concerns. The method leverages mmWave radar to generate 4-D time-series point cloud data, which is then projected onto the depth-azimuth and depth-elevation planes. This projection helps mitigate the sparsity inherent in traditional point cloud data and reduces the complexity of the machine learning model required for pose estimation. The input data structure is optimized using a sliding window technique, where consecutive frames are processed by a convolutional neural network (CNN) to extract spatial features. These features are then sorted chronologically and fed into a bi-directional long short-term memory (BiLSTM) to capture temporal features, resulting in the accurate localization of 25 skeletal joints. To validate the performance and effectiveness of the proposed method, we created a dataset comprising three body types and ten distinct actions. The experimental results demonstrate the method’s outstanding human pose estimation capability.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"11683-11696"},"PeriodicalIF":4.3,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761548","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}
This article proposes a method for generating quadrature self-mixing interference (SMI) signals using the current modulation for real-time displacement sensing. By applying a square-wave current to a semiconductor laser, periodically phase-modulated SMI signals are produced at a certain distance. Subsequently, a pair of quadrature SMI signals is obtained by separately sampling at the high and low levels of the square-wave signal, followed by an interpolation process. Finally, the target’s motion displacement is reconstructed by quadrature phase unwrapping algorithm. The method demonstrates excellent robustness, achieving reconstruction accuracy better than 20 nm under 7-dB Gaussian noise in simulation. Experiments for both sinusoidal vibrations and nonharmonic moving targets are conducted, with a displacement reconstruction error of 22.3 nm. This method does not require any additional optical components to produce phase modulation, significantly reducing system complexity and cost. Furthermore, this article analyzes the algorithm’s applicability under different distances, current modulation coefficients, and modulation voltages, providing a reference for the application of current modulation methods in SMI displacement measurements.
{"title":"Quadrature Signal Extraction by Current Modulation in Self-Mixing Interferometry for Displacement Sensing","authors":"Hanqiao Chen;Shuangxi Zhang;Xiaohan Mao;Desheng Zhu;Xiong Zheng;Zhipeng Dong;Xiulin Wang;Wencai Huang","doi":"10.1109/JSEN.2025.3543725","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3543725","url":null,"abstract":"This article proposes a method for generating quadrature self-mixing interference (SMI) signals using the current modulation for real-time displacement sensing. By applying a square-wave current to a semiconductor laser, periodically phase-modulated SMI signals are produced at a certain distance. Subsequently, a pair of quadrature SMI signals is obtained by separately sampling at the high and low levels of the square-wave signal, followed by an interpolation process. Finally, the target’s motion displacement is reconstructed by quadrature phase unwrapping algorithm. The method demonstrates excellent robustness, achieving reconstruction accuracy better than 20 nm under 7-dB Gaussian noise in simulation. Experiments for both sinusoidal vibrations and nonharmonic moving targets are conducted, with a displacement reconstruction error of 22.3 nm. This method does not require any additional optical components to produce phase modulation, significantly reducing system complexity and cost. Furthermore, this article analyzes the algorithm’s applicability under different distances, current modulation coefficients, and modulation voltages, providing a reference for the application of current modulation methods in SMI displacement measurements.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"10831-10838"},"PeriodicalIF":4.3,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761546","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}
Pub Date : 2025-02-27DOI: 10.1109/JSEN.2025.3543726
Ce Jing;Elisa Bertolesi;Guanwen Huang;Xin Li;Qin Zhang;Weiwei Zhai;Guolin Liu;Hang Li
Accelerometer and global navigation satellite system (GNSS) can be effectively combined to establish a robust multisensor deformation monitoring system. However, GNSS signals may get downgraded in challenging environments, and then destroy the Kalman filter data fusion model. As a result, the accelerometer becomes the only reliable sensor for deformation monitoring, but relying on only accelerometer data may lead to rapid error accumulation due to its potential baseline shift error. To mitigate this challenge, especially in the slow-moving deformation scenarios, we propose a baseline correction prediction algorithm named CNN-based baseline correction (CNN-BC), based on convolutional neural networks. This algorithm utilizes high-frequency acceleration and baseline correction as input and output features, respectively. The baseline correction of the training dataset is derived from the accelerometer and GNSS coupled algorithm. By incorporating the reliable prediction from the network, we can correct the original accelerometer data and reduce error accumulation. To further address the divergence in deformation velocity, we develop a convolutional neural network (CNN)-dVel, which uses high-frequency acceleration and velocity difference as input and output features, respectively. We validated the proposed algorithms through two slow deformation experiments utilizing both high-precision and low-cost accelerometers. The results demonstrate that the CNN-BC can predict reliable baseline correction, with an average root mean square (rms) of $boldsymbol {textbf {0}.textbf {37}~textbf {cm}/textbf {s}^{textbf {2}}}$ , and the CNN-dVel achieves nondivergent deformation velocity prediction, with an average rms of 0.42 cm/s. Furthermore, optimizing the training dataset with an acceleration standard deviation (STD) basis enhances prediction accuracy.
{"title":"Predicting Accelerometer Baseline Correction and Nondivergent Deformation Velocity Based on Convolutional Neural Network (CNN) During GNSS Downgrade","authors":"Ce Jing;Elisa Bertolesi;Guanwen Huang;Xin Li;Qin Zhang;Weiwei Zhai;Guolin Liu;Hang Li","doi":"10.1109/JSEN.2025.3543726","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3543726","url":null,"abstract":"Accelerometer and global navigation satellite system (GNSS) can be effectively combined to establish a robust multisensor deformation monitoring system. However, GNSS signals may get downgraded in challenging environments, and then destroy the Kalman filter data fusion model. As a result, the accelerometer becomes the only reliable sensor for deformation monitoring, but relying on only accelerometer data may lead to rapid error accumulation due to its potential baseline shift error. To mitigate this challenge, especially in the slow-moving deformation scenarios, we propose a baseline correction prediction algorithm named CNN-based baseline correction (CNN-BC), based on convolutional neural networks. This algorithm utilizes high-frequency acceleration and baseline correction as input and output features, respectively. The baseline correction of the training dataset is derived from the accelerometer and GNSS coupled algorithm. By incorporating the reliable prediction from the network, we can correct the original accelerometer data and reduce error accumulation. To further address the divergence in deformation velocity, we develop a convolutional neural network (CNN)-dVel, which uses high-frequency acceleration and velocity difference as input and output features, respectively. We validated the proposed algorithms through two slow deformation experiments utilizing both high-precision and low-cost accelerometers. The results demonstrate that the CNN-BC can predict reliable baseline correction, with an average root mean square (rms) of <inline-formula> <tex-math>$boldsymbol {textbf {0}.textbf {37}~textbf {cm}/textbf {s}^{textbf {2}}}$ </tex-math></inline-formula>, and the CNN-dVel achieves nondivergent deformation velocity prediction, with an average rms of 0.42 cm/s. Furthermore, optimizing the training dataset with an acceleration standard deviation (STD) basis enhances prediction accuracy.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"11982-11994"},"PeriodicalIF":4.3,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748780","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}
Pub Date : 2025-02-27DOI: 10.1109/JSEN.2025.3542298
Nusrat Praween;Pammi Guru Krishna Thej;Palash Kumar Basu
Glypican1 and mucin1 antigens are prominent biomarkers for the prognosis and diagnosis of pancreatic cancer. Their presence within the extracellular vesicles (EVs) opens the possibilities for oncology care through the development of minimally invasive biomarker-assisted screening tools. Traditionally, EV antigen quantification relies on ultracentrifugation (UC) and chemical lysis, which are time-consuming, equipment-dependent, and often compromise EV integrity, damaging surface intact biomarkers. This study integrates EV isolation and electric field (EF) lysis into a unified platform. The lysates were then analyzed using an electrochemical impedance spectroscopy (EIS)-based sensor to detect glypican-1 (GPC1) and mucin-1 (MUC1). ELISA confirms the EF lysis of the immobilized EV and shows an increase in the antigen concentration by 2.5 times (compared to the pre-lysed sample). Hence, EF lysis makes the sensor more sensitive than traditional methods. To enhance the electric lysis process, we applied varying voltages of a sinusoidal signal to the screen printed gold electrode (SPGE)-immobilized EVs. The lysate was subsequently used to quantify the GPC1 and MUC1 antigens through EIS. The results indicate that a 50-mV sinusoidal signal is sufficient to effectively lyse EVs, confirmed by western blotting. The nanoparticle tracking analyzer (NTA) results showed the successful isolation of $10^{{9}}$ EVs from $100~mu $ L of serum using CD63 antibody. The developed EIS sensor can detect GPC1 and MUC1 with an LOD of 0.053 and 0.033 pg/mL, respectively, from EV lysate, showing minimal nonspecific binding in the negative control. Beyond GPC1 and MUC1, the approach is adaptable for detecting other EV-associated biomarkers, enabling broader applications in early cancer detection and disease monitoring.
{"title":"Electrochemical Impedance-Based Detection of Pancreatic Cancer Biomarker Glypican1 and Mucin1 Using Electric Field-Lysed Extracellular Vesicles for Analysis: A Proof of Concept","authors":"Nusrat Praween;Pammi Guru Krishna Thej;Palash Kumar Basu","doi":"10.1109/JSEN.2025.3542298","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3542298","url":null,"abstract":"Glypican1 and mucin1 antigens are prominent biomarkers for the prognosis and diagnosis of pancreatic cancer. Their presence within the extracellular vesicles (EVs) opens the possibilities for oncology care through the development of minimally invasive biomarker-assisted screening tools. Traditionally, EV antigen quantification relies on ultracentrifugation (UC) and chemical lysis, which are time-consuming, equipment-dependent, and often compromise EV integrity, damaging surface intact biomarkers. This study integrates EV isolation and electric field (EF) lysis into a unified platform. The lysates were then analyzed using an electrochemical impedance spectroscopy (EIS)-based sensor to detect glypican-1 (GPC1) and mucin-1 (MUC1). ELISA confirms the EF lysis of the immobilized EV and shows an increase in the antigen concentration by 2.5 times (compared to the pre-lysed sample). Hence, EF lysis makes the sensor more sensitive than traditional methods. To enhance the electric lysis process, we applied varying voltages of a sinusoidal signal to the screen printed gold electrode (SPGE)-immobilized EVs. The lysate was subsequently used to quantify the GPC1 and MUC1 antigens through EIS. The results indicate that a 50-mV sinusoidal signal is sufficient to effectively lyse EVs, confirmed by western blotting. The nanoparticle tracking analyzer (NTA) results showed the successful isolation of <inline-formula> <tex-math>$10^{{9}}$ </tex-math></inline-formula> EVs from <inline-formula> <tex-math>$100~mu $ </tex-math></inline-formula>L of serum using CD63 antibody. The developed EIS sensor can detect GPC1 and MUC1 with an LOD of 0.053 and 0.033 pg/mL, respectively, from EV lysate, showing minimal nonspecific binding in the negative control. Beyond GPC1 and MUC1, the approach is adaptable for detecting other EV-associated biomarkers, enabling broader applications in early cancer detection and disease monitoring.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"10566-10574"},"PeriodicalIF":4.3,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748807","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}
Fish use their lateral line organs to sense wake changes in the flow field environment to conduct hunting, avoidance, and communication activities. Seals efficiently enhance their sensitivity to prey’s wake-induced vibrations (WIVs) by suppressing vortex-induced vibrations (VIVs) with undulating whiskers. In this study, we proposed a fully 3-D-printed piezoresistive bionic flow sensor. The bionic whisker simulates the geometrical dimensions of a real gray harbor whisker. The piezoresistive sensing unit made of liquid metal simulates the synaptic electromechanical transitions of hair follicle cells to acquire flow field information. Uniform flow field experiments and simulation results revealed that the sensor can effectively suppress VIVs at an angle of attack (AOA) of 0° and achieve a minimum flow velocity of 0.03 m/s at AOA of 90°. The sensing unit’s directional arrangement realizes the discrimination of the flow direction. In addition, the sensor demonstrated to determine the wake generated by the upstream cylinders and direct information about the upstream cylinders. Therefore, the sensors’ experimental results of the bionic whisker sensor can be applied to underwater robots to perceive diverse flow field information.
{"title":"A Fully 3-D-Printed Piezoresistive Bionic Seal Whisker Integrating Multiple Liquid Metal Tunnels for Enhanced Sensitivity in Hydrodynamic Flow Sensing","authors":"Yanbo Xu;Zengxing Zhang;Weihong Ouyang;Jiangong Cui;Xingxu Zhang;Chenyang Xue","doi":"10.1109/JSEN.2025.3543614","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3543614","url":null,"abstract":"Fish use their lateral line organs to sense wake changes in the flow field environment to conduct hunting, avoidance, and communication activities. Seals efficiently enhance their sensitivity to prey’s wake-induced vibrations (WIVs) by suppressing vortex-induced vibrations (VIVs) with undulating whiskers. In this study, we proposed a fully 3-D-printed piezoresistive bionic flow sensor. The bionic whisker simulates the geometrical dimensions of a real gray harbor whisker. The piezoresistive sensing unit made of liquid metal simulates the synaptic electromechanical transitions of hair follicle cells to acquire flow field information. Uniform flow field experiments and simulation results revealed that the sensor can effectively suppress VIVs at an angle of attack (AOA) of 0° and achieve a minimum flow velocity of 0.03 m/s at AOA of 90°. The sensing unit’s directional arrangement realizes the discrimination of the flow direction. In addition, the sensor demonstrated to determine the wake generated by the upstream cylinders and direct information about the upstream cylinders. Therefore, the sensors’ experimental results of the bionic whisker sensor can be applied to underwater robots to perceive diverse flow field information.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"10811-10819"},"PeriodicalIF":4.3,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761545","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}
Pub Date : 2025-02-26DOI: 10.1109/JSEN.2025.3542869
J. N. V. R. Swarup Kumar;Kuna Venkateswararao;Umashankar Ghugar;Sourav Kumar Bhoi;Kshira Sagar Sahoo
There has been a significant paradigm shift from wired to wireless technology in in-vehicular networks. This shift is driven by the need for greater scalability, cost-effectiveness, and flexibility. In the automotive industry, traditional wired protocols such as the local interconnect network (LIN) and media-oriented systems transport (MOST) for non-critical systems (NCSs) add complexity to installation and maintenance, incur higher material costs, and offer limited scalability and mobility. NCSs, such as infotainment and weather forecast systems, do not require low latency and do not impair vehicle function when unavailable. This article presents an advanced methodology for enhancing connectivity in noncritical in-vehicular networks using Nordic Semiconductor’s (nRFs) enhanced-nRF24L01 (EnRF24L01) module. The EnRF24L01 module is the nRF24L01 module that incorporated the Sensor-Medium Access Control (S-MAC) algorithm for energy-efficient communication. The proposed method enables seamless communication between NCSs using a tree-based primary and secondary architecture, where the primary is the actuator and the secondary is the sensor node. To optimize energy efficiency using synchronized sleep/wake schedules, reduce power consumption, and enhance scalability, the S-MAC protocol was incorporated. Comprehensive experiments were conducted in simulated environments using optimized network engineering tool (OPNET) and Proteus Circuit Simulators, analyzing critical performance metrics: latency, jitter, throughput, packet delivery ratio (PDR), and energy efficiency. The results indicate that the proposed method supports a greater number of nodes with enhanced data transmission rates and operates at lower voltages, thereby extending the communication range and reducing overall power consumption. Additionally, hardware simulation results demonstrate the successful integration of EnRF24L01 modules with Arduino for wireless data transmission, showing significant improvements in scalability, energy efficiency, and adaptability, as well as architectural and operational costs and maintenance efficiency.
{"title":"Enhanced Connectivity for Non-Critical In-Vehicle Systems Using EnRF24L01","authors":"J. N. V. R. Swarup Kumar;Kuna Venkateswararao;Umashankar Ghugar;Sourav Kumar Bhoi;Kshira Sagar Sahoo","doi":"10.1109/JSEN.2025.3542869","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3542869","url":null,"abstract":"There has been a significant paradigm shift from wired to wireless technology in in-vehicular networks. This shift is driven by the need for greater scalability, cost-effectiveness, and flexibility. In the automotive industry, traditional wired protocols such as the local interconnect network (LIN) and media-oriented systems transport (MOST) for non-critical systems (NCSs) add complexity to installation and maintenance, incur higher material costs, and offer limited scalability and mobility. NCSs, such as infotainment and weather forecast systems, do not require low latency and do not impair vehicle function when unavailable. This article presents an advanced methodology for enhancing connectivity in noncritical in-vehicular networks using Nordic Semiconductor’s (nRFs) enhanced-nRF24L01 (EnRF24L01) module. The EnRF24L01 module is the nRF24L01 module that incorporated the Sensor-Medium Access Control (S-MAC) algorithm for energy-efficient communication. The proposed method enables seamless communication between NCSs using a tree-based primary and secondary architecture, where the primary is the actuator and the secondary is the sensor node. To optimize energy efficiency using synchronized sleep/wake schedules, reduce power consumption, and enhance scalability, the S-MAC protocol was incorporated. Comprehensive experiments were conducted in simulated environments using optimized network engineering tool (OPNET) and Proteus Circuit Simulators, analyzing critical performance metrics: latency, jitter, throughput, packet delivery ratio (PDR), and energy efficiency. The results indicate that the proposed method supports a greater number of nodes with enhanced data transmission rates and operates at lower voltages, thereby extending the communication range and reducing overall power consumption. Additionally, hardware simulation results demonstrate the successful integration of EnRF24L01 modules with Arduino for wireless data transmission, showing significant improvements in scalability, energy efficiency, and adaptability, as well as architectural and operational costs and maintenance efficiency.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"11955-11962"},"PeriodicalIF":4.3,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748725","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}
In this article, we propose a novel method that integrates deep learning with Fabry-Pérot liquid crystal (FP-LC) technology for fiber Bragg grating (FBG) interrogation. The use of FP-LC enhances the measurement range and enables high sensitivity in FBG sensors, making them appropriate for a wide range of applications requiring precise and responsive sensing. However, collecting a large amount of real experimental FBG sensor data is time-consuming, technically challenging, and resource-intensive. To address this issue, we utilize a conditional generative adversarial network (CGAN) to generate a sufficient amount of synthetic training data. The CGAN generates data conditioned on real FBG sensor data, ensuring that the generated data closely look like real experimental data distributions, which is crucial for effective model training. Moreover, we proposed a convolutional neural network (CNN) method to solve crosstalk problems, to improve sensing accuracy, and to precisely detect the peak wavelength of each FBG sensor. The experimental results demonstrated that the proposed CGAN technique effectively generates a large amount of data to improve the performance of the proposed CNN model. Furthermore, the results proved that the CNN trained on CGAN-generated data significantly improves the detection speed and accuracy of central wavelength measurements compared to traditional approaches. Hence, the proposed system is cost-effective, easy to set up for experiments, increases the feasibility and portability of modularization, fast and flexible, overcoming data shortages, and improving the sensing accuracy of wavelength detection for FBG sensor systems.
{"title":"Enhanced Fiber Bragg Grating Interrogation Using Deep Learning and Fabry-Pérot Liquid Crystal: A CGAN-CNN for Improved Wavelength Detection","authors":"Minyechil Alehegn Tefera;Cheng-Kai Yao;Hao-Kuan Lee;Ssu-Han Liu;Yibeltal Chanie Manie;Ming-Che Chan;Peng-Chun Peng","doi":"10.1109/JSEN.2025.3543132","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3543132","url":null,"abstract":"In this article, we propose a novel method that integrates deep learning with Fabry-Pérot liquid crystal (FP-LC) technology for fiber Bragg grating (FBG) interrogation. The use of FP-LC enhances the measurement range and enables high sensitivity in FBG sensors, making them appropriate for a wide range of applications requiring precise and responsive sensing. However, collecting a large amount of real experimental FBG sensor data is time-consuming, technically challenging, and resource-intensive. To address this issue, we utilize a conditional generative adversarial network (CGAN) to generate a sufficient amount of synthetic training data. The CGAN generates data conditioned on real FBG sensor data, ensuring that the generated data closely look like real experimental data distributions, which is crucial for effective model training. Moreover, we proposed a convolutional neural network (CNN) method to solve crosstalk problems, to improve sensing accuracy, and to precisely detect the peak wavelength of each FBG sensor. The experimental results demonstrated that the proposed CGAN technique effectively generates a large amount of data to improve the performance of the proposed CNN model. Furthermore, the results proved that the CNN trained on CGAN-generated data significantly improves the detection speed and accuracy of central wavelength measurements compared to traditional approaches. Hence, the proposed system is cost-effective, easy to set up for experiments, increases the feasibility and portability of modularization, fast and flexible, overcoming data shortages, and improving the sensing accuracy of wavelength detection for FBG sensor systems.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"11123-11130"},"PeriodicalIF":4.3,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748741","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}
Pub Date : 2025-02-26DOI: 10.1109/JSEN.2025.3543552
Zhen Cheng;Heng Liu;Jianlong Zheng;Weihua Gong;Kaikai Chi
Mobile molecular communication (MMC) system has been widely used in the field of medical drug delivery, disease detection, and target tracking. Considering the MMC system with one mobile transmitter and multiple fixed receivers that are bionanosensors, it is of great significance to localize and track the mobile transmitter in this MMC system. The existing work mainly focused on the deep neural network (DNN), which was utilized to locate the position of the receiver in static molecular communication (MC). In this article, we consider using the Transformer-based model to predict the position of the transmitter in a 2-D unbounded MMC environment by capturing the time series of the number of molecules in each time slot at multiple receivers. The simulation results show that the Transformer-based model performs better than the DNN-based model in localizing the transmitter. When the time slot is smaller, we find that the model can approximately track the trajectory of the transmitter. In addition, we also demonstrate the factors that affect the performance of localizing and tracking the transmitter by using the model.
{"title":"Localizing and Tracking the Transmitter Bionanosensor in Mobile Molecular Communication by Deep Learning","authors":"Zhen Cheng;Heng Liu;Jianlong Zheng;Weihua Gong;Kaikai Chi","doi":"10.1109/JSEN.2025.3543552","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3543552","url":null,"abstract":"Mobile molecular communication (MMC) system has been widely used in the field of medical drug delivery, disease detection, and target tracking. Considering the MMC system with one mobile transmitter and multiple fixed receivers that are bionanosensors, it is of great significance to localize and track the mobile transmitter in this MMC system. The existing work mainly focused on the deep neural network (DNN), which was utilized to locate the position of the receiver in static molecular communication (MC). In this article, we consider using the Transformer-based model to predict the position of the transmitter in a 2-D unbounded MMC environment by capturing the time series of the number of molecules in each time slot at multiple receivers. The simulation results show that the Transformer-based model performs better than the DNN-based model in localizing the transmitter. When the time slot is smaller, we find that the model can approximately track the trajectory of the transmitter. In addition, we also demonstrate the factors that affect the performance of localizing and tracking the transmitter by using the model.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"10583-10593"},"PeriodicalIF":4.3,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748962","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}
Pub Date : 2025-02-26DOI: 10.1109/JSEN.2025.3543362
Minghua Wang;Bowen Zhao;Yi Zhang;Di Wu;Yue Wang;Xinlin Qing;Yishou Wang
Aircraft composite structures are vulnerable to barely visible impact damage (BVID) caused by external impacts. Timely identification of impact forces through sparse sensor networks is critical for structural maintenance and flight safety. This article proposes an impact force reconstruction method based on wavelet transform (WT) and low-frequency response components (LRCs). The impact forces at unknown locations can be identified using the limited training data. The method extracts LRCs via WT, ensuring stable system modeling by avoiding high-frequency disturbances. A similarity-based decision strategy adaptively selects sensor combinations and LRCs for interpolation, enabling effective impact force reconstruction through sparse networks. The approach is applicable to both reinforced and flat structural areas, offering a balanced solution between monitoring cost and reconstruction capability. Validation is provided through low-velocity impact experiments on composite stiffened panels.
{"title":"Impact Force Reconstruction in Composite Structures Using Wavelet Transform and Low-Frequency Response Components","authors":"Minghua Wang;Bowen Zhao;Yi Zhang;Di Wu;Yue Wang;Xinlin Qing;Yishou Wang","doi":"10.1109/JSEN.2025.3543362","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3543362","url":null,"abstract":"Aircraft composite structures are vulnerable to barely visible impact damage (BVID) caused by external impacts. Timely identification of impact forces through sparse sensor networks is critical for structural maintenance and flight safety. This article proposes an impact force reconstruction method based on wavelet transform (WT) and low-frequency response components (LRCs). The impact forces at unknown locations can be identified using the limited training data. The method extracts LRCs via WT, ensuring stable system modeling by avoiding high-frequency disturbances. A similarity-based decision strategy adaptively selects sensor combinations and LRCs for interpolation, enabling effective impact force reconstruction through sparse networks. The approach is applicable to both reinforced and flat structural areas, offering a balanced solution between monitoring cost and reconstruction capability. Validation is provided through low-velocity impact experiments on composite stiffened panels.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"11697-11709"},"PeriodicalIF":4.3,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761583","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}
Pub Date : 2025-02-26DOI: 10.1109/JSEN.2025.3539783
Rajeev Kumar;Lokendra Singh;S. Malathi
Human health is affected by a vast spectrum of ailments, collectively called diseases, which obstruct the proper functioning of the body and may stem from diverse origins, including genetic alterations, environmental factors, or pathogenic infections. These diseases are often categorized into groups, such as infectious diseases, triggered by bacteria, viruses, fungi, or parasites, and noninfectious diseases, encompassing genetic disorders, autoimmune conditions, and persistent illnesses like cardiovascular disease and diabetes. Infectious diseases may disseminate through direct or indirect contact, air, food, water, or vectors, whereas noninfectious diseases are frequently associated with lifestyle practices, genetic predisposition, or environmental influences. Progress in modern medicine, including vaccines, antibiotics, and diagnostic technologies, has markedly reduced the impact of numerous diseases; however, challenges continue due to the emergence of new infections, antibiotic resistance, and the escalating prevalence of noncommunicable diseases. Confronting human diseases necessitates a holistic strategy that integrates public health measures, healthcare interventions, and biomedical research to enhance prevention, diagnosis, and treatment methods, consequently improving global health outcomes. Infectious and noninfectious diseases (autoimmune disorders, cancer, and cardiovascular diseases) caused by bacteria, viruses, and fungi can be identified quickly and accurately with surface plasmon resonance (SPR) sensors. The ease and capacity of SPR technology to track biomolecular interactions enhances its adaptability, which allows early diagnosis and individualized treatment plans for a variety of medical disorders.
{"title":"Infected and Noninfected Diseases Detection for Human Health Using Surface Plasmon Resonance Biosensors: A Review","authors":"Rajeev Kumar;Lokendra Singh;S. Malathi","doi":"10.1109/JSEN.2025.3539783","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3539783","url":null,"abstract":"Human health is affected by a vast spectrum of ailments, collectively called diseases, which obstruct the proper functioning of the body and may stem from diverse origins, including genetic alterations, environmental factors, or pathogenic infections. These diseases are often categorized into groups, such as infectious diseases, triggered by bacteria, viruses, fungi, or parasites, and noninfectious diseases, encompassing genetic disorders, autoimmune conditions, and persistent illnesses like cardiovascular disease and diabetes. Infectious diseases may disseminate through direct or indirect contact, air, food, water, or vectors, whereas noninfectious diseases are frequently associated with lifestyle practices, genetic predisposition, or environmental influences. Progress in modern medicine, including vaccines, antibiotics, and diagnostic technologies, has markedly reduced the impact of numerous diseases; however, challenges continue due to the emergence of new infections, antibiotic resistance, and the escalating prevalence of noncommunicable diseases. Confronting human diseases necessitates a holistic strategy that integrates public health measures, healthcare interventions, and biomedical research to enhance prevention, diagnosis, and treatment methods, consequently improving global health outcomes. Infectious and noninfectious diseases (autoimmune disorders, cancer, and cardiovascular diseases) caused by bacteria, viruses, and fungi can be identified quickly and accurately with surface plasmon resonance (SPR) sensors. The ease and capacity of SPR technology to track biomolecular interactions enhances its adaptability, which allows early diagnosis and individualized treatment plans for a variety of medical disorders.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 7","pages":"10556-10565"},"PeriodicalIF":4.3,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748957","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：