Pub Date : 2026-01-12DOI: 10.1109/JSEN.2025.3648747
Zihan Yin;Akhilesh Jaiswal
The rapid advancement of neural network applications necessitates hardware that not only accelerates computation but also adapts efficiently to dynamic processing requirements. While processing-in-pixel has emerged as a promising solution to overcome the bottlenecks of traditional architectures at the extreme-edge, existing implementations face limitations in reconfigurability and scalability due to their static nature and inefficient area usage. Addressing these challenges, we present a novel architecture that significantly enhances the capabilities of processing-in-pixel for convolutional neural networks (CNNs). Our design innovatively integrates nonvolatile memory (NVM) with novel unit pixel circuit design, enabling dynamic reconfiguration of synaptic weights, kernel size, channel size, and stride size; thus, offering unprecedented flexibility and adaptability. By using a separate die for the pixel circuit and storing synaptic weights, our circuit achieves a substantial reduction in the required area per pixel, thereby increasing the density and scalability of the pixel array. Simulation results demonstrate dot product operations of the circuit, the nonlinearity of its analog output and a novel bucket-select curvefit model is proposed to capture it. This work not only addresses the limitations of current in-pixel computing approaches but also opens new avenues for developing more efficient, flexible, and scalable neural network hardware, paving the way for advanced artificial intelligence (AI) applications.
{"title":"FPCA: Field-Programmable Pixel Convolutional Array for Extreme-Edge Intelligence","authors":"Zihan Yin;Akhilesh Jaiswal","doi":"10.1109/JSEN.2025.3648747","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3648747","url":null,"abstract":"The rapid advancement of neural network applications necessitates hardware that not only accelerates computation but also adapts efficiently to dynamic processing requirements. While processing-in-pixel has emerged as a promising solution to overcome the bottlenecks of traditional architectures at the extreme-edge, existing implementations face limitations in reconfigurability and scalability due to their static nature and inefficient area usage. Addressing these challenges, we present a novel architecture that significantly enhances the capabilities of processing-in-pixel for convolutional neural networks (CNNs). Our design innovatively integrates nonvolatile memory (NVM) with novel unit pixel circuit design, enabling dynamic reconfiguration of synaptic weights, kernel size, channel size, and stride size; thus, offering unprecedented flexibility and adaptability. By using a separate die for the pixel circuit and storing synaptic weights, our circuit achieves a substantial reduction in the required area per pixel, thereby increasing the density and scalability of the pixel array. Simulation results demonstrate dot product operations of the circuit, the nonlinearity of its analog output and a novel bucket-select curvefit model is proposed to capture it. This work not only addresses the limitations of current in-pixel computing approaches but also opens new avenues for developing more efficient, flexible, and scalable neural network hardware, paving the way for advanced artificial intelligence (AI) applications.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 3","pages":"5254-5268"},"PeriodicalIF":4.3,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082015","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 : 2025-12-31DOI: 10.1109/JSEN.2025.3647370
Sage Lyon;Calvin Alexander Ng;Connor Pozzi;Murat Inalpolat;Christopher Niezrecki;Yan Luo
Acoustic sensors have been deployed to monitor the health of infrastructure such as wind turbine blades. Such sensors typically use wireless links to transmit sensor data; however, due to harsh, time-varying environmental conditions, these links are susceptible to interference and signal attenuation, leading to data loss. This study investigates how antenna configurations and sensor orientations can be used to address these problems. Three experimental phases were conducted in this study: 1) discrete rotation tests evaluating network performance in fixed rotational increments; 2) continuous rotation tests simulating real-world turbine operating conditions; and 3) validation using data collected from our sensing system deployed on an operational turbine. Performance metrics include received signal strength indicator (RSSI), round-trip time (RTT), and throughput. Results demonstrate that a dual-antenna sensor installed on the shear web of a turbine blade can provide reliable network performance. Using evaluations both in the laboratory and on an operational wind turbine, this work is the first to provide invaluable insights into attenuation effects from blade material and environmental interactions, aiming to optimize network performance for wireless sensors in real-world structural health monitoring.
{"title":"Signal Strength and Network Performance Optimization of a Wireless Acoustic Sensor for Wind Turbine Blade Health Monitoring","authors":"Sage Lyon;Calvin Alexander Ng;Connor Pozzi;Murat Inalpolat;Christopher Niezrecki;Yan Luo","doi":"10.1109/JSEN.2025.3647370","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3647370","url":null,"abstract":"Acoustic sensors have been deployed to monitor the health of infrastructure such as wind turbine blades. Such sensors typically use wireless links to transmit sensor data; however, due to harsh, time-varying environmental conditions, these links are susceptible to interference and signal attenuation, leading to data loss. This study investigates how antenna configurations and sensor orientations can be used to address these problems. Three experimental phases were conducted in this study: 1) discrete rotation tests evaluating network performance in fixed rotational increments; 2) continuous rotation tests simulating real-world turbine operating conditions; and 3) validation using data collected from our sensing system deployed on an operational turbine. Performance metrics include received signal strength indicator (RSSI), round-trip time (RTT), and throughput. Results demonstrate that a dual-antenna sensor installed on the shear web of a turbine blade can provide reliable network performance. Using evaluations both in the laboratory and on an operational wind turbine, this work is the first to provide invaluable insights into attenuation effects from blade material and environmental interactions, aiming to optimize network performance for wireless sensors in real-world structural health monitoring.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 3","pages":"5195-5203"},"PeriodicalIF":4.3,"publicationDate":"2025-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082162","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 : 2025-12-31DOI: 10.1109/JSEN.2025.3647487
P. V. Pravija Raj;Pranav M. Pawar;Raja Muthalagu;Ashish Gupta
The Internet of Things (IoT) is advancing at a rapid pace with wireless sensor networks (WSNs) and is becoming integral for real-time monitoring and data-driven decision-making in different sectors. Emerging real-world applications are greatly reliant on IoT-based WSNs to improve productivity and operational efficiency. However, WSNs encounter significant challenges in handling substantial data amounts and maintaining robust security requirements. These networks are prone to many security risks that may compromise network operation and data integrity. This article proposes a novel technique by integrating blockchain and incremental extreme learning machine (abbreviated as BIELM) to effectively detect such malicious nodes in WSNs and ensure secure data collection. BIELM utilizes an interplanetary file system (IPFS) for efficient data storage and blockchain to record hashes and provide secure data access. This work also introduces an effective feature selection method by modifying the sparrow search optimization and the edited nearest neighbor (ENN) for data balancing with a combination of a standard oversampling technique (SMOTE). The experimental results reveal an accuracy of 99.38%, demonstrating the superior performance of BIELM in detecting malicious nodes, with an average cost saving of approximately 22.72% in blockchain transaction costs across key operations compared to conventional blockchainbased methods, revealing its overall efficiency over the existing methods.
{"title":"BIELM: Integrating Incremental Extreme Learning Machine and Blockchain for Secure Data Gathering in IoT-Based WSNs","authors":"P. V. Pravija Raj;Pranav M. Pawar;Raja Muthalagu;Ashish Gupta","doi":"10.1109/JSEN.2025.3647487","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3647487","url":null,"abstract":"The Internet of Things (IoT) is advancing at a rapid pace with wireless sensor networks (WSNs) and is becoming integral for real-time monitoring and data-driven decision-making in different sectors. Emerging real-world applications are greatly reliant on IoT-based WSNs to improve productivity and operational efficiency. However, WSNs encounter significant challenges in handling substantial data amounts and maintaining robust security requirements. These networks are prone to many security risks that may compromise network operation and data integrity. This article proposes a novel technique by integrating blockchain and incremental extreme learning machine (abbreviated as BIELM) to effectively detect such malicious nodes in WSNs and ensure secure data collection. BIELM utilizes an interplanetary file system (IPFS) for efficient data storage and blockchain to record hashes and provide secure data access. This work also introduces an effective feature selection method by modifying the sparrow search optimization and the edited nearest neighbor (ENN) for data balancing with a combination of a standard oversampling technique (SMOTE). The experimental results reveal an accuracy of 99.38%, demonstrating the superior performance of BIELM in detecting malicious nodes, with an average cost saving of approximately 22.72% in blockchain transaction costs across key operations compared to conventional blockchainbased methods, revealing its overall efficiency over the existing methods.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 3","pages":"5204-5212"},"PeriodicalIF":4.3,"publicationDate":"2025-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082131","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 : 2025-12-30DOI: 10.1109/JSEN.2025.3646617
Zeyang Li;Chen Chen;Jie Zhang;Claudio R. C. M. da Silva;Okan Yurduseven;Trung Q. Duong;Simon L. Cotton
A new standard, IEEE 802.11bf, has been created to offer Wi-Fi sensing capabilities across sub-7 GHz and millimeter-wave bands. Applications of Wi-Fi sensing rely on channel state information (CSI) to enable various applications such as motion detection, activity recognition, and gesture recognition. Within this context, this article investigates part of the 6-GHz spectrum for use in Wi-Fi sensing, with the aim of recognizing different types of line-of-sight (LOS) perturbation. To achieve this, a novel feature extraction methodology is presented, along with innovative features designed to comprehensively capture information from CSI. More precisely, a novel random forest (RF)-based algorithm is introduced that automatically selects optimal features and constructs accurate decision trees for the classification of various human interactions with the LOS link between two Wi-Fi devices. The proposed feature extraction and selection methodology leverages variations in the channel, which are manifested by the changes in the characteristics of signal propagation caused by movements in proximity of the LOS link. Using statistical channel metrics, which can be directly linked to the physical channel, enhances the efficiency and accuracy of LOS perturbation classification. A detailed set of experiments is used to demonstrate the accuracy of our approach, which we call channel model-based features-RF (CMF-RF). CMF-RF has been shown to outperform existing methods when used to classify human interactions with the LOS link.
{"title":"Sensing Line-of-Sight Perturbations in 6-GHz Wi-Fi Using Channel Model-Based Features","authors":"Zeyang Li;Chen Chen;Jie Zhang;Claudio R. C. M. da Silva;Okan Yurduseven;Trung Q. Duong;Simon L. Cotton","doi":"10.1109/JSEN.2025.3646617","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3646617","url":null,"abstract":"A new standard, IEEE 802.11bf, has been created to offer Wi-Fi sensing capabilities across sub-7 GHz and millimeter-wave bands. Applications of Wi-Fi sensing rely on channel state information (CSI) to enable various applications such as motion detection, activity recognition, and gesture recognition. Within this context, this article investigates part of the 6-GHz spectrum for use in Wi-Fi sensing, with the aim of recognizing different types of line-of-sight (LOS) perturbation. To achieve this, a novel feature extraction methodology is presented, along with innovative features designed to comprehensively capture information from CSI. More precisely, a novel random forest (RF)-based algorithm is introduced that automatically selects optimal features and constructs accurate decision trees for the classification of various human interactions with the LOS link between two Wi-Fi devices. The proposed feature extraction and selection methodology leverages variations in the channel, which are manifested by the changes in the characteristics of signal propagation caused by movements in proximity of the LOS link. Using statistical channel metrics, which can be directly linked to the physical channel, enhances the efficiency and accuracy of LOS perturbation classification. A detailed set of experiments is used to demonstrate the accuracy of our approach, which we call channel model-based features-RF (CMF-RF). CMF-RF has been shown to outperform existing methods when used to classify human interactions with the LOS link.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 3","pages":"5181-5194"},"PeriodicalIF":4.3,"publicationDate":"2025-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082135","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 : 2025-12-30DOI: 10.1109/JSEN.2025.3645357
Satish Kumar Satti;M. Prasad
Air writing is a cutting-edge method of contactless human–machine interaction. It involves writing characters or words in the air with fingertip gestures. This method replaces keyboards and touchscreens, making it particularly useful for smart devices, healthcare applications, and handsfree text input. Predicting a single character in air writing is simple. However, detecting and classifying multiple or overlapping characters remains difficult. To address this issue, we proposed a vision-sensor-based approach that includes a Hand Tracking Algorithm and a ResYOLO-Transformer model. We also use the chaotic honey badge algorithm to optimize hyperparameters. This ensures an ideal balance across parameters. It helps avoid local optima and enhances the exploration-exploitation balance, improving prediction accuracy. A custom dataset with 26 classes was created. We used specific hand gestures to ensure that each character’s coordinates were recorded separately, even if they overlapped. The proposed model was trained and evaluated on custom and ISI datasets. It achieved an accuracy of 97.49%, demonstrating its effectiveness in robust air-written character detection and classification. Compared to other cutting-edge models such as YOLOV5, YOLOV7, YOLOV9, YOLOV11, and vision transformer (ViT), the proposed ResYOLO-Transformer model performs better. Furthermore, when integrated with the chaotic honey badger algorithm (CHBA), the proposed model outperformed other optimization techniques like CSO, PSO, BSO, and CJAYA. It achieved an improved prediction accuracy of 98.89%.
{"title":"Air-Written Multicharacter Detection and Classification Using Vision-Based Hand Gestures and an Optimized ResYOLO-Transformer","authors":"Satish Kumar Satti;M. Prasad","doi":"10.1109/JSEN.2025.3645357","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3645357","url":null,"abstract":"Air writing is a cutting-edge method of contactless human–machine interaction. It involves writing characters or words in the air with fingertip gestures. This method replaces keyboards and touchscreens, making it particularly useful for smart devices, healthcare applications, and handsfree text input. Predicting a single character in air writing is simple. However, detecting and classifying multiple or overlapping characters remains difficult. To address this issue, we proposed a vision-sensor-based approach that includes a Hand Tracking Algorithm and a ResYOLO-Transformer model. We also use the chaotic honey badge algorithm to optimize hyperparameters. This ensures an ideal balance across parameters. It helps avoid local optima and enhances the exploration-exploitation balance, improving prediction accuracy. A custom dataset with 26 classes was created. We used specific hand gestures to ensure that each character’s coordinates were recorded separately, even if they overlapped. The proposed model was trained and evaluated on custom and ISI datasets. It achieved an accuracy of 97.49%, demonstrating its effectiveness in robust air-written character detection and classification. Compared to other cutting-edge models such as YOLOV5, YOLOV7, YOLOV9, YOLOV11, and vision transformer (ViT), the proposed ResYOLO-Transformer model performs better. Furthermore, when integrated with the chaotic honey badger algorithm (CHBA), the proposed model outperformed other optimization techniques like CSO, PSO, BSO, and CJAYA. It achieved an improved prediction accuracy of 98.89%.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 3","pages":"5229-5240"},"PeriodicalIF":4.3,"publicationDate":"2025-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082051","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 recent years, the widespread adoption of drones, while offering convenience, has also led to significant security challenges such as illegal intrusions and privacy violations, creating an urgent need for reliable identification and classification systems. A primary obstacle to achieving this reliability is the high similarity of radio frequency (RF) signals among different drone models, which often leads to misclassification. In this study, we propose the DS-UAVNet, a network that employs a dual-branch architecture to independently process complementary information from the time and frequency domains, thereby preventing information loss. Within this network, a designed parallel convolution module efficiently extracts multiscale features while reducing model complexity. To address the inherent vulnerabilities of the single-modality drone identification system, we further design M-DS-UAVNet, a multimodal framework that enhances identification robustness by leveraging a transfer learning strategy to fuse audio and RF features. Evaluations show that DS-UAVNet achieves accuracies of 98.74% and 98.56% on the public DroneRF dataset for drone classification and operation mode recognition, respectively, outperforming existing methods. Moreover, the M-DS-UAVNet framework achieves 100.00% and 99.78% accuracy on the constructed multimodal dataset, validating the effectiveness of the multimodal fusion strategy for building identification systems.
{"title":"An Efficient Dual-Branch Network and Multimodal Fusion Framework for Drone Identification","authors":"Borong Fu;Yan Zhang;Jiaming Wu;Feiyang Ye;Wancheng Zhang","doi":"10.1109/JSEN.2025.3645409","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3645409","url":null,"abstract":"In recent years, the widespread adoption of drones, while offering convenience, has also led to significant security challenges such as illegal intrusions and privacy violations, creating an urgent need for reliable identification and classification systems. A primary obstacle to achieving this reliability is the high similarity of radio frequency (RF) signals among different drone models, which often leads to misclassification. In this study, we propose the DS-UAVNet, a network that employs a dual-branch architecture to independently process complementary information from the time and frequency domains, thereby preventing information loss. Within this network, a designed parallel convolution module efficiently extracts multiscale features while reducing model complexity. To address the inherent vulnerabilities of the single-modality drone identification system, we further design M-DS-UAVNet, a multimodal framework that enhances identification robustness by leveraging a transfer learning strategy to fuse audio and RF features. Evaluations show that DS-UAVNet achieves accuracies of 98.74% and 98.56% on the public DroneRF dataset for drone classification and operation mode recognition, respectively, outperforming existing methods. Moreover, the M-DS-UAVNet framework achieves 100.00% and 99.78% accuracy on the constructed multimodal dataset, validating the effectiveness of the multimodal fusion strategy for building identification systems.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 3","pages":"5241-5253"},"PeriodicalIF":4.3,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082011","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}
To address the synergistic challenges of energy efficiency and security in wireless sensor networks (WSNs) under complex attack environments, this article proposes a collaborative energy efficiency and security enhancement (CEESE) routing algorithm. During the clustering phase, an enhanced particle swarm optimization (PSO) method is introduced, which integrates the golden sine algorithm (Gold- SA) and Levy flight (LF) to balance the global exploration and local exploitation through sine perturbations and random long-hop mechanisms. A multiobjective fitness function is constructed, considering node residual energy, comprehensive trust values, and communication distance, thereby achieving the energy-balanced cluster head (CH) election. In the routing phase, a hybrid trust model combining direct and indirect trust is designed, which collaborates with deep Q (DQ)-learning to enable real-time path state awareness and dynamic maintenance. Simulation results demonstrate that CEESE achieves the superior performance across varying network scales and attack scenarios. Specifically, in a 100-node, 100 × 100 m monitoring area, the first node death round of CEESE improved by 37.8%, 36.9%, 14.7%, 10.4%, and 7.1% compared with TAOSC-MHR, MRCH, EEHCHR, DST-WOA, and CTRF algorithms, respectively. Its advantages persist in a large-scale network with 500 nodes within a 200 × 200 m area. Regarding security, under a black-hole attack involving 50% malicious nodes, CEESE achieves a packet delivery rate (PDR) 19.2%–59.3% higher, a malicious node detection rate (DR) 5.7%–28.1% higher, and an average delay 14.3%–46.5% lower than the compared algorithms. This study provides an efficient routing solution for energy-constrained WSN applications with stringent security requirements.
{"title":"Collaborative Energy Efficiency and Security Enhancement Routing Algorithm Based on Enhanced PSO and DQ-Learning for WSNs","authors":"Liubao Zhang;Cuiran Li;Jiahui Xu;Li Liu;Jianli Xie","doi":"10.1109/JSEN.2025.3643463","DOIUrl":"https://doi.org/10.1109/JSEN.2025.3643463","url":null,"abstract":"To address the synergistic challenges of energy efficiency and security in wireless sensor networks (WSNs) under complex attack environments, this article proposes a collaborative energy efficiency and security enhancement (CEESE) routing algorithm. During the clustering phase, an enhanced particle swarm optimization (PSO) method is introduced, which integrates the golden sine algorithm (Gold- SA) and Levy flight (LF) to balance the global exploration and local exploitation through sine perturbations and random long-hop mechanisms. A multiobjective fitness function is constructed, considering node residual energy, comprehensive trust values, and communication distance, thereby achieving the energy-balanced cluster head (CH) election. In the routing phase, a hybrid trust model combining direct and indirect trust is designed, which collaborates with deep Q (DQ)-learning to enable real-time path state awareness and dynamic maintenance. Simulation results demonstrate that CEESE achieves the superior performance across varying network scales and attack scenarios. Specifically, in a 100-node, 100 × 100 m monitoring area, the first node death round of CEESE improved by 37.8%, 36.9%, 14.7%, 10.4%, and 7.1% compared with TAOSC-MHR, MRCH, EEHCHR, DST-WOA, and CTRF algorithms, respectively. Its advantages persist in a large-scale network with 500 nodes within a 200 × 200 m area. Regarding security, under a black-hole attack involving 50% malicious nodes, CEESE achieves a packet delivery rate (PDR) 19.2%–59.3% higher, a malicious node detection rate (DR) 5.7%–28.1% higher, and an average delay 14.3%–46.5% lower than the compared algorithms. This study provides an efficient routing solution for energy-constrained WSN applications with stringent security requirements.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 3","pages":"5165-5180"},"PeriodicalIF":4.3,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082049","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：