Junjie Zhao
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
Junjie Zhao has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Sen Qiao
School of Electronics and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China
5 shared publications - Jinwei Wang
Nankai University, Tianjin, China
4 shared publications - Guangjie Liu
Nanjing University of Information Science and Technology
3 shared publications - Huihui Ding
School of Electronics and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China
3 shared publications - Xiaopeng Ji
School of Electronic and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China
3 shared publications - James Msughter Adeke
School of Electronics and Information Engineering, Nanjing University of Information Science and Technology, Nanjing, 210044, China
2 shared publications
Latest Publications
- Open Access
Optimization of Covert Communication in Multi-Sensor Asymmetric Noise Systems
This work investigates wireless covert communication in a multi-sensor asymmetric noise scenario. We adopt KL (Kullback–Leibler) divergence as the covertness constraint metric and mutual information as the transmission rate metric. To accurately approximate KL divergence and mutual information in covert communication, we employ the Taylor series expansion technique. Analytical expressions for KL divergence and mutual information in covert communication are derived, and we optimize the amplitude gain and phase angles based on these analytical expressions. Our findings underscore the importance of phase angle selection in covert communication within asymmetric noise systems. We propose an effective method for optimizing the transmission amplitude gain and phase angles in scenarios with asymmetric noise. Numerical results validate the effectiveness and superiority of our proposed method.
Open AccessOptimization of Covert Communication in Multi-Sensor Asymmetric Noise Systems
This work investigates wireless covert communication in a multi-sensor asymmetric noise scenario. We adopt KL (Kullback–Leibler) divergence as the covertness constraint metric and mutual information as the transmission rate metric. To accurately approximate KL divergence and mutual information in covert communication, we employ the Taylor series expansion technique. Analytical expressions for KL divergence and mutual information in covert communication are derived, and we optimize the amplitude gain and phase angles based on these analytical expressions. Our findings underscore the importance of phase angle selection in covert communication within asymmetric noise systems. We propose an effective method for optimizing the transmission amplitude gain and phase angles in scenarios with asymmetric noise. Numerical results validate the effectiveness and superiority of our proposed method.
- Open Access
- 19 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Generating Image Adversarial Example by Modifying JPEG Stream
Research on adversarial attacks mainly focuses on reducing the amplitude of disturbances, increasing the success rate of attacks, and improving attack efficiency. However, the adversarial examples are all in the form of raw (float matrix or lossless compressed pictures). Due to the difficulty of storage and transmission caused by a large amount of redundant information, raw pictures rarely exist in the real world. In this work, a convolution operation is used to simulate the JPEG decompression process, forming a decompression module independent of the classification model to obtain the gradient information about the JPEG stream. We establish a process for generating JPEG adversarial examples. To improve the decoded image’s visual quality and reduce the perturbation amplitude, we select the complex texture area according to the human visual model and select only the most effective points in each DCT block to embed adversarial noise. A large number of tests on the ImageNet dataset show the effectiveness of the proposed method.
- 19 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessGenerating Image Adversarial Example by Modifying JPEG Stream
Research on adversarial attacks mainly focuses on reducing the amplitude of disturbances, increasing the success rate of attacks, and improving attack efficiency. However, the adversarial examples are all in the form of raw (float matrix or lossless compressed pictures). Due to the difficulty of storage and transmission caused by a large amount of redundant information, raw pictures rarely exist in the real world. In this work, a convolution operation is used to simulate the JPEG decompression process, forming a decompression module independent of the classification model to obtain the gradient information about the JPEG stream. We establish a process for generating JPEG adversarial examples. To improve the decoded image’s visual quality and reduce the perturbation amplitude, we select the complex texture area according to the human visual model and select only the most effective points in each DCT block to embed adversarial noise. A large number of tests on the ImageNet dataset show the effectiveness of the proposed method.
- Open Access
- 9 January 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (2) , 577
Covert Surveillance Video Transmission with QAM Constellations Modulation
This study explores the covert transmission of surveillance videos using QAM (quadrature amplitude modulation) constellations. By analyzing transmission strategies used in surveillance video systems, we adopt KL divergence as the constraint for covertness performance and mutual information to characterize transmission rates. Utilizing the Taylor series expansion method, we simplify the complex integral calculations of mutual information and KL divergence into summation operations. This simplification not only reduces computational complexity but also ensures precise calculations. Theoretical derivations establish the necessary conditions for the covert transmission of surveillance videos, optimizing modulation probabilities for constellation points. Experimental results and simulation validations demonstrate the superior performance of our proposed method in terms of both information transfer and precision.
- 9 January 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (2) , 577
Open AccessCovert Surveillance Video Transmission with QAM Constellations Modulation
This study explores the covert transmission of surveillance videos using QAM (quadrature amplitude modulation) constellations. By analyzing transmission strategies used in surveillance video systems, we adopt KL divergence as the constraint for covertness performance and mutual information to characterize transmission rates. Utilizing the Taylor series expansion method, we simplify the complex integral calculations of mutual information and KL divergence into summation operations. This simplification not only reduces computational complexity but also ensures precise calculations. Theoretical derivations establish the necessary conditions for the covert transmission of surveillance videos, optimizing modulation probabilities for constellation points. Experimental results and simulation validations demonstrate the superior performance of our proposed method in terms of both information transfer and precision.
- Open Access
- 5 January 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (2) , 483
The Optimization of Covert Communication in Asymmetric Jammer-Assist Systems
This research investigates the feasibility of covert and reliable communication between Alice and Bob in the presence of a vigilant adversary, Willie, employing a jammer in a complex Gaussian channel. Alice strategically manipulates the noise-power ratio along the real and imaginary axes to disrupt Willie’s detection capabilities. Covertness constraints are quantified using KL divergence, and transmission performance is assessed through mutual information. Additionally, an optimization method is introduced to enhance covert communication by fine-tuning the amplitude gain and noise-power ratio. Numerical results confirm the effectiveness and superiority of the proposed method, showcasing its ability to maximize covert transmission rates.
- 5 January 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (2) , 483
Open AccessThe Optimization of Covert Communication in Asymmetric Jammer-Assist Systems
This research investigates the feasibility of covert and reliable communication between Alice and Bob in the presence of a vigilant adversary, Willie, employing a jammer in a complex Gaussian channel. Alice strategically manipulates the noise-power ratio along the real and imaginary axes to disrupt Willie’s detection capabilities. Covertness constraints are quantified using KL divergence, and transmission performance is assessed through mutual information. Additionally, an optimization method is introduced to enhance covert communication by fine-tuning the amplitude gain and noise-power ratio. Numerical results confirm the effectiveness and superiority of the proposed method, showcasing its ability to maximize covert transmission rates.
- Open Access
- 16 December 2023
- Published by MDPI AG in Future Internet
- Vol. 15 (12) , 405
Securing Network Traffic Classification Models against Adversarial Examples Using Derived Variables
Machine learning (ML) models are essential to securing communication networks. However, these models are vulnerable to adversarial examples (AEs), in which malicious inputs are modified by adversaries to produce the desired output. Adversarial training is an effective defense method against such attacks but relies on access to a substantial number of AEs, a prerequisite that entails significant computational resources and the inherent limitation of poor performance on clean data. To address these problems, this study proposes a novel approach to improve the robustness of ML-based network traffic classification models by integrating derived variables (DVars) into training. Unlike adversarial training, our approach focuses on enhancing training using DVars, introducing randomness into the input data. DVars are generated from the baseline dataset and significantly improve the resilience of the model to AEs. To evaluate the effectiveness of DVars, experiments were conducted using the CSE-CIC-IDS2018 dataset and three state-of-the-art ML-based models: decision tree (DT), random forest (RF), and k-neighbors (KNN). The results show that DVars can improve the accuracy of KNN under attack from 0.45% to 0.84% for low-intensity attacks and from 0.32% to 0.66% for high-intensity attacks. Furthermore, both DT and RF achieve a significant increase in accuracy when subjected to attack of different intensity. Moreover, DVars are computationally efficient, scalable, and do not require access to AEs.
- 16 December 2023
- Published by MDPI AG in Future Internet
- Vol. 15 (12) , 405
Open AccessSecuring Network Traffic Classification Models against Adversarial Examples Using Derived Variables
Machine learning (ML) models are essential to securing communication networks. However, these models are vulnerable to adversarial examples (AEs), in which malicious inputs are modified by adversaries to produce the desired output. Adversarial training is an effective defense method against such attacks but relies on access to a substantial number of AEs, a prerequisite that entails significant computational resources and the inherent limitation of poor performance on clean data. To address these problems, this study proposes a novel approach to improve the robustness of ML-based network traffic classification models by integrating derived variables (DVars) into training. Unlike adversarial training, our approach focuses on enhancing training using DVars, introducing randomness into the input data. DVars are generated from the baseline dataset and significantly improve the resilience of the model to AEs. To evaluate the effectiveness of DVars, experiments were conducted using the CSE-CIC-IDS2018 dataset and three state-of-the-art ML-based models: decision tree (DT), random forest (RF), and k-neighbors (KNN). The results show that DVars can improve the accuracy of KNN under attack from 0.45% to 0.84% for low-intensity attacks and from 0.32% to 0.66% for high-intensity attacks. Furthermore, both DT and RF achieve a significant increase in accuracy when subjected to attack of different intensity. Moreover, DVars are computationally efficient, scalable, and do not require access to AEs.
- 27 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 14, 248-253
Neural network-based adaptive sliding mode control for multi-joint manipulator
Considering the problems of slow convergence speed and low tracking accuracy in the trajectory tracking control system of the multi-DOF manipulator, it is of great significance to design an effective control method to quickly track the target trajectories. In this paper, the dynamic modeling of the manipulator is conducted, followed by the development of a control strategy and an analysis of its stability. Firstly, the Lagrange method is applied to establish dynamic modeling of the manipulator. Secondly, a controller based on neural network adaptive sliding mode control is proposed to realize the trajectory tracking of the manipulator, which realizes the suppression of external disturbance and overcomes the influence of friction. Then, the Lyapunov direct method was applied to prove the stability of the designed system to guarantee the convergence of the tracking error. Finally, the experimental results based on Simulink under software Matlab indicate that the designed control strategy absolutely solve the problems of slow convergence speed and low tracking accuracy. It can adaptively compensate the uncertainties of the system and has good robustness against external interference.
- 27 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 14, 248-253
Neural network-based adaptive sliding mode control for multi-joint manipulator
Considering the problems of slow convergence speed and low tracking accuracy in the trajectory tracking control system of the multi-DOF manipulator, it is of great significance to design an effective control method to quickly track the target trajectories. In this paper, the dynamic modeling of the manipulator is conducted, followed by the development of a control strategy and an analysis of its stability. Firstly, the Lagrange method is applied to establish dynamic modeling of the manipulator. Secondly, a controller based on neural network adaptive sliding mode control is proposed to realize the trajectory tracking of the manipulator, which realizes the suppression of external disturbance and overcomes the influence of friction. Then, the Lyapunov direct method was applied to prove the stability of the designed system to guarantee the convergence of the tracking error. Finally, the experimental results based on Simulink under software Matlab indicate that the designed control strategy absolutely solve the problems of slow convergence speed and low tracking accuracy. It can adaptively compensate the uncertainties of the system and has good robustness against external interference.
- 12 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 37, 1135-1140
Hidden Markov model based finite-time H∞ guaranteed cost control for singular discrete-time Markov jump delay systems
Considering partially known transition rates and saturating actuators, the finite-time H ∞ control problem for a class of singular discrete-time Markovian jump delayed systems is studied in this article. First, by employing local sector conditions and an appropriate Lyapunov function, a hidden Markov model based state feedback controller is designed to guarantee that the resulted closed-loop constrained system is mean-square locally finite-time stabilizable, and the closed-loop cost function value is not more than a specified upper bound. Furthermore, some sufficient conditions for the solution to this problem are derived in terms of linear matrix inequalities. Finally, a numerical example is provided to demonstrate the effectiveness of proposed method.
- 12 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 37, 1135-1140
Hidden Markov model based finite-time H∞ guaranteed cost control for singular discrete-time Markov jump delay systems
Considering partially known transition rates and saturating actuators, the finite-time H ∞ control problem for a class of singular discrete-time Markovian jump delayed systems is studied in this article. First, by employing local sector conditions and an appropriate Lyapunov function, a hidden Markov model based state feedback controller is designed to guarantee that the resulted closed-loop constrained system is mean-square locally finite-time stabilizable, and the closed-loop cost function value is not more than a specified upper bound. Furthermore, some sufficient conditions for the solution to this problem are derived in terms of linear matrix inequalities. Finally, a numerical example is provided to demonstrate the effectiveness of proposed method.
- Open Access
- 30 April 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (9) , 2379
Detecting High-Resolution Adversarial Images with Few-Shot Deep Learning
Deep learning models have enabled significant performance improvements to remote sensing image processing. Usually, a large number of training samples is required for detection models. In this study, a dynamic simulation training strategy is designed to generate samples in real time during training. The few adversarial examples are not only directly involved in the training but are also used to fit the distribution model of adversarial noise, helping the real-time generated samples to be similar to adversarial examples. The noise of the training samples is randomly generated according to the distribution model, and the random variation of training inputs reduces the overfitting phenomenon. To enhance the detectability of adversarial noise, the input model is no longer a normalized image but a JPEG error image. Experiments show that with the proposed dynamic simulation training strategy, common classification models such as ResNet and DenseNet can effectively detect adversarial examples.
- 30 April 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (9) , 2379
Open AccessDetecting High-Resolution Adversarial Images with Few-Shot Deep Learning
Deep learning models have enabled significant performance improvements to remote sensing image processing. Usually, a large number of training samples is required for detection models. In this study, a dynamic simulation training strategy is designed to generate samples in real time during training. The few adversarial examples are not only directly involved in the training but are also used to fit the distribution model of adversarial noise, helping the real-time generated samples to be similar to adversarial examples. The noise of the training samples is randomly generated according to the distribution model, and the random variation of training inputs reduces the overfitting phenomenon. To enhance the detectability of adversarial noise, the input model is no longer a normalized image but a JPEG error image. Experiments show that with the proposed dynamic simulation training strategy, common classification models such as ResNet and DenseNet can effectively detect adversarial examples.
- 26 May 2022
- Published by Springer Nature in Multimedia Systems
- Vol. 28 (5) , 1809-1822
ESGAN for generating high quality enhanced samples
- 26 May 2022
- Published by Springer Nature in Multimedia Systems
- Vol. 28 (5) , 1809-1822
ESGAN for generating high quality enhanced samples
- 8 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 24 (15209210) , 230-244
SmsNet: A New Deep Convolutional Neural Network Model for Adversarial Example Detection
The emergence of adversarial examples has had a significant impact on the development and application of deep learning. In this paper, a novel convolutional neural network model, the stochastic multifilter statistical network (SmsNet), is proposed for the detection of adversarial examples. A feature statistical layer is constructed to collect statistical data of feature map output from each convolutional layer in SmsNet by combining manual features with a neural network. The entire model is an end-to-end detection model, so the feature statistical layer is not independent of the network, and its output is directly transmitted to the fully connected layer by a short-cut connection called the SmsConnection. Additionally, a dynamic pruning strategy is introduced to simplify the model structure for better performance. The experiments demonstrate the effectiveness of the network structure and pruning strategy, and the proposed model achieves high detection rates against state-of-the-art adversarial attacks.
- 8 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 24 (15209210) , 230-244
SmsNet: A New Deep Convolutional Neural Network Model for Adversarial Example Detection
The emergence of adversarial examples has had a significant impact on the development and application of deep learning. In this paper, a novel convolutional neural network model, the stochastic multifilter statistical network (SmsNet), is proposed for the detection of adversarial examples. A feature statistical layer is constructed to collect statistical data of feature map output from each convolutional layer in SmsNet by combining manual features with a neural network. The entire model is an end-to-end detection model, so the feature statistical layer is not independent of the network, and its output is directly transmitted to the fully connected layer by a short-cut connection called the SmsConnection. Additionally, a dynamic pruning strategy is introduced to simplify the model structure for better performance. The experiments demonstrate the effectiveness of the network structure and pruning strategy, and the proposed model achieves high detection rates against state-of-the-art adversarial attacks.