Naike Du
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Naike Du has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Xiuzhu Ye
School of Information and Electronics Engineering, Beijing Institute of Technology, Beijing 100081, China
6 shared publications - Guangzhong Zhang
School of Information and Electronics Engineering, Beijing Institute of Technology, Beijing 100081, China
3 shared publications - Yuchao Guo
School of Information and Electronics Engineering, Beijing Institute of Technology, Beijing, China
3 shared publications - Rencheng Song
Hefei University of Technology
2 shared publications - Jing Wang
School of Information and Electronics, Beijing Institute of Technology, Beijing, China
1 shared publication - Jinyang Liu
School of Information and Electronics, Beijing Institute of Technology, Beijing, China
1 shared publication
Latest Publications
- Open Access
- 28 October 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (21) , 3704
Motion-Adaptive Through-Wall Moving Target Enhancement Based on Energy-Guided Competitive Fusion
Reliable through-wall human imaging is challenging because strong wall clutter and motion-dependent target responses can degrade target preservation, particularly under low signal-to-noise ratio (SNR) conditions. This paper proposes a motion-adaptive through-wall moving-target enhancement framework based on energy-guided competitive fusion. After wall clutter suppression and wall-compensated imaging, each pixelwise complex slow-time sequence is decomposed into complementary zero-Doppler and nonzero-Doppler response branches. Kapur maximum-entropy segmentation is then used to construct a common candidate support, from which a data-derived three-branch energy reference distribution is obtained for the zero-Doppler, nonzero-Doppler, and virtual-background hypotheses. The resulting reference-weighted Softmax fusion is formulated as the closed-form solution of a Kullback–Leibler (KL)-regularized optimization problem. Simulations and measured through-wall experiments show that the two decomposed branches provide different relative responses to targets with different motion states, while the proposed fusion improves overall target-background separation.
- 28 October 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (21) , 3704
Open AccessMotion-Adaptive Through-Wall Moving Target Enhancement Based on Energy-Guided Competitive Fusion
Reliable through-wall human imaging is challenging because strong wall clutter and motion-dependent target responses can degrade target preservation, particularly under low signal-to-noise ratio (SNR) conditions. This paper proposes a motion-adaptive through-wall moving-target enhancement framework based on energy-guided competitive fusion. After wall clutter suppression and wall-compensated imaging, each pixelwise complex slow-time sequence is decomposed into complementary zero-Doppler and nonzero-Doppler response branches. Kapur maximum-entropy segmentation is then used to construct a common candidate support, from which a data-derived three-branch energy reference distribution is obtained for the zero-Doppler, nonzero-Doppler, and virtual-background hypotheses. The resulting reference-weighted Softmax fusion is formulated as the closed-form solution of a Kullback–Leibler (KL)-regularized optimization problem. Simulations and measured through-wall experiments show that the two decomposed branches provide different relative responses to targets with different motion states, while the proposed fusion improves overall target-background separation.
- 16 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-17
Real-Time Through-Wall Detection of Stationary Human Targets Using Micro-Doppler and Adaptive Interference Suppression
Through-wall radar (TWR) has significant potential in security surveillance and post-disaster search and rescue. However, in complex environments, weak human echoes are often masked by strong clutter and multi-source interference, severely degrading detection performance. This paper presents a novel detection method that integrates micro-Doppler analysis with adaptive interference suppression for stationary human targets. For narrowband interference, an improved vector mean-canceling algorithm (VMCA) combined with range–time two-dimensional filtering is introduced to achieve adaptive clutter and interference suppression. For broadband interference, a frequency-domain adaptive filtering technique is developed that leverages the complex-baseband architecture to construct a negative-frequency reference channel from the conjugate correlation between positive and negative spectra, thereby canceling broadband noise and pushing the measurement sensitivity of TWR towards its hardware noise limit. To enable real-time deployment, an efficient parallel pipelined processing architecture is implemented on a Xilinx XC7Z035 field-programmable gate array (FPGA), with total power consumption below 3W and a core processing delay of approximately 10ms. Experimental results on a through-wall radar testbed demonstrate that, compared with representative recent matrix decomposition methods, the proposed framework provides improved interference suppression and more stable micro-Doppler energy focusing across diverse target postures, leading to consistently higher detection performance in cluttered scenarios. These results confirm the effectiveness and practical engineering feasibility of the proposed framework in complex through-wall environments.
- 16 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-17
Real-Time Through-Wall Detection of Stationary Human Targets Using Micro-Doppler and Adaptive Interference Suppression
Through-wall radar (TWR) has significant potential in security surveillance and post-disaster search and rescue. However, in complex environments, weak human echoes are often masked by strong clutter and multi-source interference, severely degrading detection performance. This paper presents a novel detection method that integrates micro-Doppler analysis with adaptive interference suppression for stationary human targets. For narrowband interference, an improved vector mean-canceling algorithm (VMCA) combined with range–time two-dimensional filtering is introduced to achieve adaptive clutter and interference suppression. For broadband interference, a frequency-domain adaptive filtering technique is developed that leverages the complex-baseband architecture to construct a negative-frequency reference channel from the conjugate correlation between positive and negative spectra, thereby canceling broadband noise and pushing the measurement sensitivity of TWR towards its hardware noise limit. To enable real-time deployment, an efficient parallel pipelined processing architecture is implemented on a Xilinx XC7Z035 field-programmable gate array (FPGA), with total power consumption below 3W and a core processing delay of approximately 10ms. Experimental results on a through-wall radar testbed demonstrate that, compared with representative recent matrix decomposition methods, the proposed framework provides improved interference suppression and more stable micro-Doppler energy focusing across diverse target postures, leading to consistently higher detection performance in cluttered scenarios. These results confirm the effectiveness and practical engineering feasibility of the proposed framework in complex through-wall environments.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 33, 3536-3540
Adaptive Kinematic Masking for Urban Human Activity Recognition Using UWB Radar: A Sim2Real Approach
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 33, 3536-3540
Adaptive Kinematic Masking for Urban Human Activity Recognition Using UWB Radar: A Sim2Real Approach
- Open Access
- 10 June 2023
- Published by MDPI AG in Electronics
- Vol. 12 (12) , 2623
A Machine Learning-Assisted Inversion Method for Solving Biomedical Imaging Based on Semi-Experimental Data
Machine learning approaches have been extensively utilized in the field of inverse scattering problems. Typically, the training dataset is generated synthetically using ideal radiation sources such as plane waves or cylindrical waves. However, the testing data often consist of experimental data that take into account the antenna port couplings and waveform distortions within the system. While noise can be artificially added to synthetic data, it may not accurately represent the real experimental noise. Consequently, the application of machine learning-assisted inversion techniques may encounter challenges when the training dataset differs significantly from the experimental data. In this paper, we propose an experimental system specifically designed for human body imaging. A semi-experimental training dataset is constructed using full-wave simulation software, incorporating the relative permittivities of common human tissues. Furthermore, the system noise is meticulously considered through full-wave simulation, enhancing the authenticity of the dataset. A back-propagation scheme is firstly employed to obtain the rough reconstructed images. Then, the U-net convolutional neural network (CNN) is employed to map these rough images to high-resolution ones. Numerical results demonstrate that, in comparison to networks trained solely on synthetic data, the network trained using semi-experimental data achieves superior reconstruction results with lower errors and improved image quality.
- 10 June 2023
- Published by MDPI AG in Electronics
- Vol. 12 (12) , 2623
Open AccessA Machine Learning-Assisted Inversion Method for Solving Biomedical Imaging Based on Semi-Experimental Data
Machine learning approaches have been extensively utilized in the field of inverse scattering problems. Typically, the training dataset is generated synthetically using ideal radiation sources such as plane waves or cylindrical waves. However, the testing data often consist of experimental data that take into account the antenna port couplings and waveform distortions within the system. While noise can be artificially added to synthetic data, it may not accurately represent the real experimental noise. Consequently, the application of machine learning-assisted inversion techniques may encounter challenges when the training dataset differs significantly from the experimental data. In this paper, we propose an experimental system specifically designed for human body imaging. A semi-experimental training dataset is constructed using full-wave simulation software, incorporating the relative permittivities of common human tissues. Furthermore, the system noise is meticulously considered through full-wave simulation, enhancing the authenticity of the dataset. A back-propagation scheme is firstly employed to obtain the rough reconstructed images. Then, the U-net convolutional neural network (CNN) is employed to map these rough images to high-resolution ones. Numerical results demonstrate that, in comparison to networks trained solely on synthetic data, the network trained using semi-experimental data achieves superior reconstruction results with lower errors and improved image quality.
- 28 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-2
Improvement of Vivaldi Antenna in Microwave Imaging System
Vivaldi antenna has been widely used in the microwave imaging system due to its wide frequency range and low cross-polarization character. The directivity and transfer efficiency of the antenna in the system is of great importance to the imaging quality. The Vivaldi antenna is commonly connected to the coaxial line with an SMA connector. However, the SMA connector forms a coplanar waveguide structure on the substrate, which makes the center frequency shift and the return loss increase. Through simulation, it is found that removing the SMA connector pins can improve the efficiency of the system antenna at the working frequency. The simulation results are verified by actual measurements in the imaging system.
- 28 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-2
Improvement of Vivaldi Antenna in Microwave Imaging System
Vivaldi antenna has been widely used in the microwave imaging system due to its wide frequency range and low cross-polarization character. The directivity and transfer efficiency of the antenna in the system is of great importance to the imaging quality. The Vivaldi antenna is commonly connected to the coaxial line with an SMA connector. However, the SMA connector forms a coplanar waveguide structure on the substrate, which makes the center frequency shift and the return loss increase. Through simulation, it is found that removing the SMA connector pins can improve the efficiency of the system antenna at the working frequency. The simulation results are verified by actual measurements in the imaging system.