Zhaoxi Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zhaoxi Chen has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
- Gang Li
Department of Electronic Engineering, Tsinghua University, Beijing, China
4 shared publications - Boyuan Dong
Department of Electronic Engineering, Tsinghua University, Beijing, China
2 shared publications - Francesco Fioranelli
Microwave Sensing, Signals and Systems (MS3) group, Delft University of Technology, Delft, The Netherlands
2 shared publications - Chengfang Ren
SONDRA, CentraleSuplec, Universit Paris-Saclay, Gif-sur-Yvette, France
1 shared publication - Giovanni Manfredi
Department of Information Engineering, Università Politecnica delle Marche, Ancona, Italy
1 shared publication - Huaiyuan Liang
Beihang University, Beijing, China
1 shared publication
Latest Publications
- 29 July 2025
- Published by Elsevier in Sleep Health
- Vol. 11 (6) , 859-867
Deep learning-based automated diagnosis of obstructive sleep apnea and sleep stage classification in children using millimeter-wave radar and pulse oximeter
- 29 July 2025
- Published by Elsevier in Sleep Health
- Vol. 11 (6) , 859-867
Deep learning-based automated diagnosis of obstructive sleep apnea and sleep stage classification in children using millimeter-wave radar and pulse oximeter
- 4 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 20 (1545598X) , 1-5
Azimuth Ambiguity Suppression for SAR via Variable PRF and Complex Image Deconvolution
The azimuth ambiguity phenomenon has widely existed in synthetic aperture radar (SAR). The false targets introduced by the ambiguity phenomenon may seriously affect the interpretation of SAR images. It is difficult for existing methods to suppress the azimuth ambiguity effectively at low oversampling rates. In this paper, we propose an approach of azimuth ambiguity suppression for SAR. The transmitted pulse sequence is changed to a non-uniform sampling pattern with multiple pulse repetition frequency (PRF) cycling. By taking the structural similarity of the ambiguity features as reference information, the Richardson-Lucy (RL) algorithm is utilized to suppress the ambiguity of SAR images through complex image deconvolution. The proposed method only requires simple changes to the transmitted pulse sequence and can be implemented without changing the hardware of existing SAR systems. It offers the capability to suppress the azimuth ambiguity of SAR at low oversampling rates. Experiments on real measured SAR data have verified the feasibility of the proposed method.
- 4 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 20 (1545598X) , 1-5
Azimuth Ambiguity Suppression for SAR via Variable PRF and Complex Image Deconvolution
The azimuth ambiguity phenomenon has widely existed in synthetic aperture radar (SAR). The false targets introduced by the ambiguity phenomenon may seriously affect the interpretation of SAR images. It is difficult for existing methods to suppress the azimuth ambiguity effectively at low oversampling rates. In this paper, we propose an approach of azimuth ambiguity suppression for SAR. The transmitted pulse sequence is changed to a non-uniform sampling pattern with multiple pulse repetition frequency (PRF) cycling. By taking the structural similarity of the ambiguity features as reference information, the Richardson-Lucy (RL) algorithm is utilized to suppress the ambiguity of SAR images through complex image deconvolution. The proposed method only requires simple changes to the transmitted pulse sequence and can be implemented without changing the hardware of existing SAR systems. It offers the capability to suppress the azimuth ambiguity of SAR at low oversampling rates. Experiments on real measured SAR data have verified the feasibility of the proposed method.
- 30 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 27 (4) , 1813-1824
The Human Activity Radar Challenge: Benchmarking Based on the ‘Radar Signatures of Human Activities’ Dataset From Glasgow University
Radar is an extremely valuable sensing technology for detecting moving targets and measuring their range, velocity, and angular positions. When people are monitored at home, radar is more likely to be accepted by end-users, as they already use WiFi, is perceived as privacy-preserving compared to cameras, and does not require user compliance as wearable sensors do. Furthermore, it is not affected by lighting condi-tions nor requires artificial lights that could cause discomfort in the home environment. So, radar-based human activities classification in the context of assisted living can empower an aging society to live at home independently longer. However, challenges remain as to the formulation of the most effective algorithms for radar-based human activities classification and their validation. To promote the exploration and cross-evaluation of different algorithms, our dataset released in 2019 was used to benchmark various classification approaches. The challenge was open from February 2020 to December 2020. A total of 23 organizations worldwide, forming 12 teams from academia and industry, participated in the inaugural Radar Challenge, and submitted 188 valid entries to the challenge. This paper presents an overview and evaluation of the approaches used for all primary contributions in this inaugural challenge. The proposed algorithms are summarized, and the main parameters affecting their performances are analyzed.
- 30 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 27 (4) , 1813-1824
The Human Activity Radar Challenge: Benchmarking Based on the ‘Radar Signatures of Human Activities’ Dataset From Glasgow University
Radar is an extremely valuable sensing technology for detecting moving targets and measuring their range, velocity, and angular positions. When people are monitored at home, radar is more likely to be accepted by end-users, as they already use WiFi, is perceived as privacy-preserving compared to cameras, and does not require user compliance as wearable sensors do. Furthermore, it is not affected by lighting condi-tions nor requires artificial lights that could cause discomfort in the home environment. So, radar-based human activities classification in the context of assisted living can empower an aging society to live at home independently longer. However, challenges remain as to the formulation of the most effective algorithms for radar-based human activities classification and their validation. To promote the exploration and cross-evaluation of different algorithms, our dataset released in 2019 was used to benchmark various classification approaches. The challenge was open from February 2020 to December 2020. A total of 23 organizations worldwide, forming 12 teams from academia and industry, participated in the inaugural Radar Challenge, and submitted 188 valid entries to the challenge. This paper presents an overview and evaluation of the approaches used for all primary contributions in this inaugural challenge. The proposed algorithms are summarized, and the main parameters affecting their performances are analyzed.
- 11 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
Unambiguous Doppler Extension for FMCW Radar via Poisson Disk Sampling
The maximum unambiguous Doppler in classical saw-tooth frequency-modulated continuous wave (FMCW) radar is inherently limited by the sweep repetition frequency (SRF). In this letter, an unambiguous Doppler extension method for FMCW radar is proposed, based on the combination of Poisson disk sampling (PDS) and sparse recovery. The PDS is adopted in the slow time domain such that there are random intervals between adjacent transmitted sweeps and no overlaps. Taking advantages of the excellent spectrum antialiasing performance of PDS, the maximum unambiguous Doppler of FMCW radar can be extended without reducing the range resolution. Then, the iterative soft thresholding-like (IST-like) algorithm is utilized to reconstruct the accurate range-Doppler spectrum. Compared with the existing unambiguous Doppler extension methods, the proposed method performs better in multitarget and low signal-to-noise ratio (SNR) scenarios. The effectiveness of the proposed method is verified by the experiments on real FMCW radar data.
- 11 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
Unambiguous Doppler Extension for FMCW Radar via Poisson Disk Sampling
The maximum unambiguous Doppler in classical saw-tooth frequency-modulated continuous wave (FMCW) radar is inherently limited by the sweep repetition frequency (SRF). In this letter, an unambiguous Doppler extension method for FMCW radar is proposed, based on the combination of Poisson disk sampling (PDS) and sparse recovery. The PDS is adopted in the slow time domain such that there are random intervals between adjacent transmitted sweeps and no overlaps. Taking advantages of the excellent spectrum antialiasing performance of PDS, the maximum unambiguous Doppler of FMCW radar can be extended without reducing the range resolution. Then, the iterative soft thresholding-like (IST-like) algorithm is utilized to reconstruct the accurate range-Doppler spectrum. Compared with the existing unambiguous Doppler extension methods, the proposed method performs better in multitarget and low signal-to-noise ratio (SNR) scenarios. The effectiveness of the proposed method is verified by the experiments on real FMCW radar data.
- 6 March 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 15 (5) , 669-673
Personnel Recognition and Gait Classification Based on Multistatic Micro-Doppler Signatures Using Deep Convolutional Neural Networks
In this letter, we propose two methods for personnel recognition and gait classification using deep convolutional neural networks (DCNNs) based on multistatic radar micro-Doppler signatures. Previous DCNN-based schemes have mainly focused on monostatic scenarios, whereas directional diversity offered by multistatic radar is exploited in this letter to improve classification accuracy. We first propose the voted monostatic DCNN (VMo-DCNN) method, which trains DCNNs on each receiver node separately and fuses the results by binary voting. By merging the fusion step into the network architecture, we further propose the multistatic DCNN (Mul-DCNN) method, which performs slightly better than VMo-DCNN. These methods are validated on real data measured with a 2.4-GHz multistatic radar system. Experimental results show that the Mul-DCNN achieves over 99% accuracy in armed/unarmed gait classification using only 20% training data and similar performance in two-class personnel recognition using 50% training data, which are higher than the accuracy obtained by performing DCNN on a single radar node.
- 6 March 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 15 (5) , 669-673
Personnel Recognition and Gait Classification Based on Multistatic Micro-Doppler Signatures Using Deep Convolutional Neural Networks
In this letter, we propose two methods for personnel recognition and gait classification using deep convolutional neural networks (DCNNs) based on multistatic radar micro-Doppler signatures. Previous DCNN-based schemes have mainly focused on monostatic scenarios, whereas directional diversity offered by multistatic radar is exploited in this letter to improve classification accuracy. We first propose the voted monostatic DCNN (VMo-DCNN) method, which trains DCNNs on each receiver node separately and fuses the results by binary voting. By merging the fusion step into the network architecture, we further propose the multistatic DCNN (Mul-DCNN) method, which performs slightly better than VMo-DCNN. These methods are validated on real data measured with a 2.4-GHz multistatic radar system. Experimental results show that the Mul-DCNN achieves over 99% accuracy in armed/unarmed gait classification using only 20% training data and similar performance in two-class personnel recognition using 50% training data, which are higher than the accuracy obtained by performing DCNN on a single radar node.