Xiaoying Qiu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Xiaoying Qiu has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
Jinwei Yu
School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China
6 shared publicationsWenbao Jiang
College of Computer Science, Beijing Information Science and Technology University, Beijing, China
6 shared publicationsZhaozhong Guo
College of Computer Science, Beijing Information Science and Technology University, Beijing, China
6 shared publicationsMaozhi Xu
School of Mathematical Sciences, Peking University, Beijing, China
5 shared publications- Xiaoyu Ma
Beijing Information Science and Technology University
4 shared publications - Xuan Sun
Beijing Information Science and Technology University
4 shared publications
Latest Publications
- 26 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (10) , 22045-22057
Blockchain-Assisted Meta-Learning for Efficient and Reliable Authentication in IoV
Considering the rapid variations in a frequently changing environment and the broadcast nature of wireless signals, security authentication plays a crucial role in ensuring the security of Internet of Vehicles (IoV) communications. Existing authentication schemes in IoV suffer from two main limitations: first, protocol-based authentication causes high latency and heavy cryptographic overhead when dealing with numerous concurrent requests. Second, although edge computing-assisted methods reduce cloud communication latency, the ability to support model adaptation across environments remains limited. To address these challenges, a blockchain-assisted meta-learning (BAML) authentication scheme is proposed for providing security authentication protection in IoV. Specifically, a meta-learning algorithm is designed to collaboratively learn the device/location/channel-related feature of vehicles from source environments and then to verify their identities. During the movement of vehicles, a blockchain module is proposed to guide the new environment to inherit the most optimal authentication knowledge from the meta-learning knowledge base and achieve high performance even with limited data. The BAML scheme is designed for reliable collaboration and continuous protection in IoV. Extensive experiments are conducted to validate the effectiveness of BAML. When given 9 samples, the BAML of new environments reaches 99.7%, surpassing other competitive baselines.
- 26 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (10) , 22045-22057
Blockchain-Assisted Meta-Learning for Efficient and Reliable Authentication in IoV
Considering the rapid variations in a frequently changing environment and the broadcast nature of wireless signals, security authentication plays a crucial role in ensuring the security of Internet of Vehicles (IoV) communications. Existing authentication schemes in IoV suffer from two main limitations: first, protocol-based authentication causes high latency and heavy cryptographic overhead when dealing with numerous concurrent requests. Second, although edge computing-assisted methods reduce cloud communication latency, the ability to support model adaptation across environments remains limited. To address these challenges, a blockchain-assisted meta-learning (BAML) authentication scheme is proposed for providing security authentication protection in IoV. Specifically, a meta-learning algorithm is designed to collaboratively learn the device/location/channel-related feature of vehicles from source environments and then to verify their identities. During the movement of vehicles, a blockchain module is proposed to guide the new environment to inherit the most optimal authentication knowledge from the meta-learning knowledge base and achieve high performance even with limited data. The BAML scheme is designed for reliable collaboration and continuous protection in IoV. Extensive experiments are conducted to validate the effectiveness of BAML. When given 9 samples, the BAML of new environments reaches 99.7%, surpassing other competitive baselines.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Wireless Communications Letters
- Vol. 15, 4653-4657
GPT-Inspired Adaptive Physical-Layer Authentication for Industrial Wireless Communication Systems
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Wireless Communications Letters
- Vol. 15, 4653-4657
GPT-Inspired Adaptive Physical-Layer Authentication for Industrial Wireless Communication Systems
- 17 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 30 (10897798) , 637-641
Deep Complex Network Architecture for Multi-User Physical Layer Authentication in Wireless Communication
Physical Layer Authentication (PLA) is a promising strategy for wireless security. Most existing PLA schemes have relied on real-valued neural networks, where complex-valued channel impulse response (CIR) is processed by separating the real and imaginary components into dual-channel inputs. This conversion disrupts the inherent coupling between magnitude and phase, thereby constraining authentication accuracy. Importantly, the spatial position of each user inherently serves as a reliable identity fingerprint. In this letter, a complex-valued network-based multi-task learning (CVN-MTL) model is proposed for multi-user authentication. By leveraging the spatiotemporal characteristics of both CIR and position, the CVN-MTL model simultaneously performs user authentication and fine-grained localization. Experiment results show that the CVN-MTL model performs superiority on authentication performance and is robust to different communication scenarios.
- 17 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 30 (10897798) , 637-641
Deep Complex Network Architecture for Multi-User Physical Layer Authentication in Wireless Communication
Physical Layer Authentication (PLA) is a promising strategy for wireless security. Most existing PLA schemes have relied on real-valued neural networks, where complex-valued channel impulse response (CIR) is processed by separating the real and imaginary components into dual-channel inputs. This conversion disrupts the inherent coupling between magnitude and phase, thereby constraining authentication accuracy. Importantly, the spatial position of each user inherently serves as a reliable identity fingerprint. In this letter, a complex-valued network-based multi-task learning (CVN-MTL) model is proposed for multi-user authentication. By leveraging the spatiotemporal characteristics of both CIR and position, the CVN-MTL model simultaneously performs user authentication and fine-grained localization. Experiment results show that the CVN-MTL model performs superiority on authentication performance and is robust to different communication scenarios.
- 4 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (22) , 48597-48609
DSDA-IRNet: Inverted Residual Network-Based IIoT Authentication With Double-Parameter Smoothing Data Augmentation Algorithm
Given the frequently changing and potentially unreliable environment, the cost-effective authentication is essential to achieve security Industrial Internet of the Things (IIoT) environment with dramatically enhanced communication and transportation safety. Although great success has been achieved for multi-identity authentication schemes, it depends on adequate data collection, which is particularly laborious and time-consuming, being impractical for actual IIoT applications or privacy sensitive environments. Moreover, authentication schemes based on deep learning may suffer from high complexity and excessive latency, leading to potential interruption of critical services in dynamic IIoT environments. To overcome the above challenges, this article proposes a lightweight and robust authentication scheme, namely, DSDA-IRNet, which combines inverted residual network (IRNet) and double smoothing data augmentation (DSDA) algorithm. Specifically, IRNet is proposed to achieve a balance between authentication complexity and accuracy. The DSDA significantly alleviates the overfitting and low authentication accuracy caused by scarce training data. Finally, extensive experimental evaluations based on industrial scenarios datasets are conducted to assess the detection performance and robustness of the DSDA-IRNet. Compared with the existing schemes, our results characterize the outperformance of the DSDA-IRNet in authentication accuracy and computational complexity.
- 4 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (22) , 48597-48609
DSDA-IRNet: Inverted Residual Network-Based IIoT Authentication With Double-Parameter Smoothing Data Augmentation Algorithm
Given the frequently changing and potentially unreliable environment, the cost-effective authentication is essential to achieve security Industrial Internet of the Things (IIoT) environment with dramatically enhanced communication and transportation safety. Although great success has been achieved for multi-identity authentication schemes, it depends on adequate data collection, which is particularly laborious and time-consuming, being impractical for actual IIoT applications or privacy sensitive environments. Moreover, authentication schemes based on deep learning may suffer from high complexity and excessive latency, leading to potential interruption of critical services in dynamic IIoT environments. To overcome the above challenges, this article proposes a lightweight and robust authentication scheme, namely, DSDA-IRNet, which combines inverted residual network (IRNet) and double smoothing data augmentation (DSDA) algorithm. Specifically, IRNet is proposed to achieve a balance between authentication complexity and accuracy. The DSDA significantly alleviates the overfitting and low authentication accuracy caused by scarce training data. Finally, extensive experimental evaluations based on industrial scenarios datasets are conducted to assess the detection performance and robustness of the DSDA-IRNet. Compared with the existing schemes, our results characterize the outperformance of the DSDA-IRNet in authentication accuracy and computational complexity.
- Open Access
- 1 January 2025
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 83 (2) , 2025-2040
A Lightweight Convolutional Neural Network with Squeeze and Excitation Module for Security Authentication Using Wireless Channel
Physical layer authentication (PLA) in the context of the Internet of Things (IoT) has gained significant attention. Compared with traditional encryption and blockchain technologies, PLA provides a more computationally efficient alternative to exploiting the properties of the wireless medium itself. Some existing PLA solutions rely on static mechanisms, which are insufficient to address the authentication challenges in fifth generation (5G) and beyond wireless networks. Additionally, with the massive increase in mobile device access, the communication security of the IoT is vulnerable to spoofing attacks. To overcome the above challenges, this paper proposes a lightweight deep convolutional neural network (CNN) equipped with squeeze and excitation module (SE module) in dynamic wireless environments, namely SE-ConvNet. To be more specific, a convolution factorization is developed to reduce the complexity of PLA models based on deep learning. Moreover, an SE module is designed in the deep CNN to enhance useful features and maximize authentication accuracy. Compared with the existing solutions, the proposed SE-ConvNet enabled PLA scheme performs excellently in mobile and time-varying wireless environments while maintaining lower computational complexity.
- 1 January 2025
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 83 (2) , 2025-2040
Open AccessA Lightweight Convolutional Neural Network with Squeeze and Excitation Module for Security Authentication Using Wireless Channel
Physical layer authentication (PLA) in the context of the Internet of Things (IoT) has gained significant attention. Compared with traditional encryption and blockchain technologies, PLA provides a more computationally efficient alternative to exploiting the properties of the wireless medium itself. Some existing PLA solutions rely on static mechanisms, which are insufficient to address the authentication challenges in fifth generation (5G) and beyond wireless networks. Additionally, with the massive increase in mobile device access, the communication security of the IoT is vulnerable to spoofing attacks. To overcome the above challenges, this paper proposes a lightweight deep convolutional neural network (CNN) equipped with squeeze and excitation module (SE module) in dynamic wireless environments, namely SE-ConvNet. To be more specific, a convolution factorization is developed to reduce the complexity of PLA models based on deep learning. Moreover, an SE module is designed in the deep CNN to enhance useful features and maximize authentication accuracy. Compared with the existing solutions, the proposed SE-ConvNet enabled PLA scheme performs excellently in mobile and time-varying wireless environments while maintaining lower computational complexity.
- Open Access
- 28 November 2024
- Published by MDPI AG in Electronics
- Vol. 13 (23) , 4699
Hierarchical Security Authentication with Attention-Enhanced Convolutional Network for Internet of Things
As security authentication issues continue to arise in future wireless communication networks, researchers are working hard to further improve authentication techniques. Recently, physical layer authentication (PLA) has received widespread attention for its lightweight nature compared to traditional encryption methods based on keys and blockchain. However, the existing PLA mechanisms based on a fixed decision threshold have low reliability in dynamic environments. Moreover, PLA solutions are typically based on binary authentication, and these binary-type schemes cannot provide different levels of access control. To address these challenges, this article introduces the concept of hierarchical security authentication, aiming to achieve multi-level secure authorization access. In order to further improve the accuracy of identity verification, we design an Attention-Enhanced Convolutional Network (AECN) model that integrates the attention mechanism. Specifically, by introducing a confidence score branch, the proposed AECN-based PLA scheme completes authentication without a threshold, thus avoiding the issues stemming from inappropriate threshold settings in conventional PLA schemes. The simulation results show that our proposed AECN framework outperforms existing algorithms at different levels of security authentication.
- 28 November 2024
- Published by MDPI AG in Electronics
- Vol. 13 (23) , 4699
Open AccessHierarchical Security Authentication with Attention-Enhanced Convolutional Network for Internet of Things
As security authentication issues continue to arise in future wireless communication networks, researchers are working hard to further improve authentication techniques. Recently, physical layer authentication (PLA) has received widespread attention for its lightweight nature compared to traditional encryption methods based on keys and blockchain. However, the existing PLA mechanisms based on a fixed decision threshold have low reliability in dynamic environments. Moreover, PLA solutions are typically based on binary authentication, and these binary-type schemes cannot provide different levels of access control. To address these challenges, this article introduces the concept of hierarchical security authentication, aiming to achieve multi-level secure authorization access. In order to further improve the accuracy of identity verification, we design an Attention-Enhanced Convolutional Network (AECN) model that integrates the attention mechanism. Specifically, by introducing a confidence score branch, the proposed AECN-based PLA scheme completes authentication without a threshold, thus avoiding the issues stemming from inappropriate threshold settings in conventional PLA schemes. The simulation results show that our proposed AECN framework outperforms existing algorithms at different levels of security authentication.
- Open Access
- 20 April 2024
- Published by MDPI AG in Electronics
- Vol. 13 (8) , 1577
Intelligent Security Authentication for Connected and Autonomous Vehicles: Attacks and Defenses
The emergence of integrated positioning, communication, and sensing technologies has paved the way for a surge in connected and autonomous vehicles. The control system has been successful in reliable and fast transmission. However, practical applications face security risks, especially data tampering and spoofing attacks. To improve the resilience of the system against potential attacks, we attempt to leverage a generative adversarial network learning-assisted authentication framework (GAF). In addition to proposing a new method for validating vehicles, we also introduce a new architectural innovation in the generator–discriminator pair to achieve improved results. The generator sub-network is constructed using an advanced convolutional neural network, whereas the discriminator is designed to leverage global and local information to determine whether a signal is real or fake. On this basis, we propose a signal enhancement-based authentication method, a deep convolutional generative adversarial network (DCGAN). Experimental results using the National Institute of Standards and Technology (NIST) dataset show that the proposed method is effective in denoising and improving the detection performance.
- 20 April 2024
- Published by MDPI AG in Electronics
- Vol. 13 (8) , 1577
Open AccessIntelligent Security Authentication for Connected and Autonomous Vehicles: Attacks and Defenses
The emergence of integrated positioning, communication, and sensing technologies has paved the way for a surge in connected and autonomous vehicles. The control system has been successful in reliable and fast transmission. However, practical applications face security risks, especially data tampering and spoofing attacks. To improve the resilience of the system against potential attacks, we attempt to leverage a generative adversarial network learning-assisted authentication framework (GAF). In addition to proposing a new method for validating vehicles, we also introduce a new architectural innovation in the generator–discriminator pair to achieve improved results. The generator sub-network is constructed using an advanced convolutional neural network, whereas the discriminator is designed to leverage global and local information to determine whether a signal is real or fake. On this basis, we propose a signal enhancement-based authentication method, a deep convolutional generative adversarial network (DCGAN). Experimental results using the National Institute of Standards and Technology (NIST) dataset show that the proposed method is effective in denoising and improving the detection performance.
- Open Access
Channel Prediction-Based Security Authentication for Artificial Intelligence of Things
The emerging physical-layer unclonable attribute-aided authentication (PLUA) schemes are capable of outperforming traditional isolated approaches, with the advantage of having reliable fingerprints. However, conventional PLUA methods face new challenges in artificial intelligence of things (AIoT) applications owing to their limited flexibility. These challenges arise from the distributed nature of AIoT devices and the involved information, as well as the requirement for short end-to-end latency. To address these challenges, we propose a security authentication scheme that utilizes intelligent prediction mechanisms to detect spoofing attack. Our approach is based on a dynamic authentication method using long short term memory (LSTM), where the edge computing node observes and predicts the time-varying channel information of access devices to detect clone nodes. Additionally, we introduce a Savitzky–Golay filter-assisted high order cumulant feature extraction model (SGF-HOCM) for preprocessing channel information. By utilizing future channel attributes instead of relying solely on previous channel information, our proposed approach enables authentication decisions. We have conducted extensive experiments in actual industrial environments to validate our prediction-based security strategy, which has achieved an accuracy of 97%.
Open AccessChannel Prediction-Based Security Authentication for Artificial Intelligence of Things
The emerging physical-layer unclonable attribute-aided authentication (PLUA) schemes are capable of outperforming traditional isolated approaches, with the advantage of having reliable fingerprints. However, conventional PLUA methods face new challenges in artificial intelligence of things (AIoT) applications owing to their limited flexibility. These challenges arise from the distributed nature of AIoT devices and the involved information, as well as the requirement for short end-to-end latency. To address these challenges, we propose a security authentication scheme that utilizes intelligent prediction mechanisms to detect spoofing attack. Our approach is based on a dynamic authentication method using long short term memory (LSTM), where the edge computing node observes and predicts the time-varying channel information of access devices to detect clone nodes. Additionally, we introduce a Savitzky–Golay filter-assisted high order cumulant feature extraction model (SGF-HOCM) for preprocessing channel information. By utilizing future channel attributes instead of relying solely on previous channel information, our proposed approach enables authentication decisions. We have conducted extensive experiments in actual industrial environments to validate our prediction-based security strategy, which has achieved an accuracy of 97%.
- Open Access
Enhanced Security Authentication Based on Convolutional-LSTM Networks
The performance of classical security authentication models can be severely affected by imperfect channel estimation as well as time-varying communication links. The commonly used approach of statistical decisions for the physical layer authenticator faces significant challenges in a dynamically changing, non-stationary environment. To address this problem, this paper introduces a deep learning-based authentication approach to learn and track the variations of channel characteristics, and thus improving the adaptability and convergence of the physical layer authentication. Specifically, an intelligent detection framework based on a Convolutional-Long Short-Term Memory (Convolutional-LSTM) network is designed to deal with channel differences without knowing the statistical properties of the channel. Both the robustness and the detection performance of the learning authentication scheme are analyzed, and extensive simulations and experiments show that the detection accuracy in time-varying environments is significantly improved.
Open AccessEnhanced Security Authentication Based on Convolutional-LSTM Networks
The performance of classical security authentication models can be severely affected by imperfect channel estimation as well as time-varying communication links. The commonly used approach of statistical decisions for the physical layer authenticator faces significant challenges in a dynamically changing, non-stationary environment. To address this problem, this paper introduces a deep learning-based authentication approach to learn and track the variations of channel characteristics, and thus improving the adaptability and convergence of the physical layer authentication. Specifically, an intelligent detection framework based on a Convolutional-Long Short-Term Memory (Convolutional-LSTM) network is designed to deal with channel differences without knowing the statistical properties of the channel. Both the robustness and the detection performance of the learning authentication scheme are analyzed, and extensive simulations and experiments show that the detection accuracy in time-varying environments is significantly improved.