ZSZhongsen Sun

Zhongsen Sun

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Zhongsen Sun has a total of 52 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Compact Circularly Polarized CRLH Leaky‐Wave Antenna for Satellite Communications

    by Xiongan Zeng, Naiqiao Zhao, Leonid F. Chernogor, Tian Liu, Zhejun Jin, Yu Zheng, Zhongsen Sun

    A circularly polarized (CP) microstrip composite right/left‐handed (CRLH) leaky‐wave antenna (LWA) with wide‐angle scanning capability is proposed. The antenna is based on a microstrip transmission line that integrates patch, metallic via, and folded branches for miniaturization. The left‐handed and right‐handed parameters generated among the branches, patch, and via are balanced, thereby eliminating the OSB effect. A prototype was fabricated and tested, and the unit cell features a compact size of 0.79 λg × 0.45 λg. Measurement results indicate that the antenna achieves a wide scanning range from −68° to +34° across 9.5–12.5 GHz (27.3%). The antenna achieves a circularly polarized beam‐scanning range from −30° to +10° over 10.3–12.0 GHz (15.2%), and the maximum gain reaches 15.2 dBi with smooth variation over the operating band. These characteristics render the proposed antenna well suited for satellite communication applications.

  • An improved real-time detection model for small targets in UAV aerial photography based on YOLOv11s

    by Yan Gao, Min Zhang, Chengyang Tian, Runce Zhang, Zhejun Jin, Zhongsen Sun, Tian Liu, Yu Zheng

    This paper addresses the challenge of small object detection in complex backgrounds under limited computational resources in unmanned aerial vehicle (UAV) aerial imagery. Based on the lightweight YOLOv11s framework, we proposes a lightweight model LMS-YOLO designed for UAV small object detection. With a view to alleviate the loss of subtle features caused by deep downsampling, we propose a local–global decoupled attention module, which splits the original global attention into local window attention and global context modules to process local details and global information respectively. Meanwhile, in order to expand the receptive field to cope with drastic scale changes, this paper proposes a multi-scale sparse dilated attention module, which can expand the receptive field without adding extra parameters through parallel processing of different dilation rates, and better capture the context around small objects. To enhance gradient sensitivity in bounding box regression and reduce missed detections of small targets, this paper proposes a loss function sensitive to small-scale objects (SS-ShapeIoU). The proposed loss function addresses low-resolution regression by penalizing edge offsets and adapting area weights. Consequently, bounding box accuracy improves significantly. Compared to the baseline YOLOv11s, LMS-YOLO improves mAP 50 by 4.37% on VisDrone-DET2019, and the precision and recall are increased by 5.60% and 2.16% respectively. Additionally, our model reduces parameters by 1.91% and FLOPs by 0.83%, while sustaining an inference speed of 32.8 FPS. Evaluations on the UAVDT dataset further validate the cross-dataset adaptability of our model. Despite varying illumination and weather conditions, its mAP 50 still improves by 2.49% over the baseline. In summary, LMS-YOLO achieves a good balance between accuracy and efficiency, offering strong practical viability for resource-constrained UAV edge devices.

  • DSP‐Reduced Reconfigurable Processor Design for Memory‐Based FFT

    by Yifan Zheng, Shuoyang Wang, Yongqiang Luo, Yu Zheng, Tian Liu, Zhongsen Sun

    The implementation flexibility of mixed‐radix fast Fourier transform (FFT) enables dynamic radix combinations for multi‐standard communication systems. However, conventional mixed‐radix FFT designs exhibit computational resource redundancy when simultaneously supporting butterfly units of multiple radices, which results in suboptimal hardware resource utilization. This paper proposes a reconfigurable mixed‐radix FFT processor incorporating an innovative reconfigurable butterfly unit design supporting radix‐2/3/5 FFT operations and a multi‐path parallel conflict‐free access method. The redesigned butterfly unit enhances hardware resource efficiency through functionally heterogeneous architecture while eliminating access conflicts. Implemented and verified on a Xilinx XC7Z020CLG400‐2 FPGA platform, comprehensive testing of the FFT processor system demonstrates a 25% reduction in digital signal processor (DSP) consumption and a 26% improvement in computational efficiency versus state‐of‐the‐art memory‐based FFT processors.

  • Optimization algorithm for DOA estimation accuracy of coherent signal sources based on sparse Bayesian theory

    by Zhuqing Mei, Xue Li, Hengfeng Li, Shuoyang Wang, Yu Zheng, Tian Liu, Zhongsen Sun

  • X-Band and Low-RCS Flexible Wideband Antenna Array Based on Metasurface

    by Hongzhi Chen, Qiutong Zhen, Jing Chen, Haoyuan Sun, Haoran Li, Leonid F. Chernogor, Zhongsen Sun, Yu Zheng, Tian Liu, Zhejun Jin

    Transparent metasurface-loaded flexible antenna for the X-band. The dumbbell-shaped metasurface is a type of polarization conversion unit arranged in a checkerboard pattern. This structure achieves 10 dB Linear polarization radar crosssection(RCS) reduction under vertical incidence within the 9.2- 21.2 GHz range. Attach it to a slotted antenna array powered by a 1-to-4 Wilkinson power divider. The antenna has good radiation characteristics within the bandwidth of the X band. This antenna design boasts a low profile (0.14λ0), flexibility, and interference resistance. In the 0-90° central angle bending scenario, the structure maintains good RCS reduction and antenna radiation properties, while also effectively reducing electromagnetic waves at 0-45° angle, making it highly suitable for wider applications such as detection radar, satellite communications, and medical equipment.


  • A Transparent and Flexible Polarization Conversion Electromagnetic Scatterer

    by Qiutong Zhen, Zhoutao Mao, Hongzhi Chen, Jing Chen, Leonid F. Chernogor, Zhejun Jin, Zhongsen Sun, Yu Zheng, Peiheng Zhou, Tian Liu

    This letter presents a novel design for a transparent and flexible electromagnetic scatterer aimed at reducing radar cross-section (RCS). Made from low-resistance ITO thin film, PET, and PVC plastic, it achieves up to 81.3% transparency. Innovative unit design overcomes ITO's conductivity limitations and broadens bandwidth. Using the Pancharatnam-Berry phase principle, rotating the top-layer conductive structure creates encoded supercells. The letter explores RCS optimization using multiple algorithms, analyzes RCS changes in MSS with ITO patches in the conformal state, and compares the effects of ITO and copper on RCS performance. Multi-algorithm optimization results in over 10 dB RCS reduction from 8.5 to 19.8 GHz, even under 2π curvature or 45° oblique incidence. With a compact size (0.135 λ 0 ), transparency(81.3%), flexibility(central angle up to 2π), and wideband performance (80%), the scatterer is ideal for applications like multispectral scattering, antenna radomes, and transparent windows.


  • A Broadband Circularly Polarized Antenna Based on Transparent Conformal Metasurface

    by Tian Liu, Lu Liu, Hongzhi Chen, Haoyuan Sun, Zhejun Jin, Leonid F. Chernogor, Dmitriy O. Batrakov, Zhongsen Sun

    A circularly polarized antenna base on the transparent conformal polarization-conversion metasurface is proposed. The designed meta-surface unit is an ellipse rotating 45 degrees with a rectangular slot. The three metasurface cells are arranged in a group and rotated 90 degrees around the feed antenna. The metasurface is printed on transparent and conformal PET while also possessing low-profile characteristics. By using this method, polarization conversion of the antenna is achieved without damaging the original antenna structure, thus extending the circular polarization bandwidth of the antenna. The simulated and measured results show that the -10 dB bandwidth of the antenna is 3.9-4.7 GHz, the axial ratio is less than 3 dB. The meta-surface effectively expands the circular polarization bandwidth of the antenna from 100 MHz to 800 MHz. The radiation efficiency of the antenna exceeds 60%, and the peak gain is more than 5 dBi. The radiation patterns show that the antenna has right-handed polarization characteristics and obvious directional radiation characteristics. The meta-surface design effectively broadens the circular polarization bandwidth of the antenna, making it suitable for circularly polarized communication systems operating in the C-band frequency range.


  • Dual-Band Dual-Polarized Antenna Based on TDLS Metasurface for 4G/5G Base Station Applications

    by Yangchao Chen, Leonid F. Chernogor, Yu Zheng, Zhejun Jin, Zhongsen Sun, Zhao Yao, Cong Wang, Tian Liu

    This letter presents a dual-band dual polarization (DP) antenna based on a cross-shaped metasurface with a terminal double-loop structure (TDLS) applied to 4G/5G base stations. The Fabry-Perot (FP) cavity is formed by introducing the TDLS metasurface above the dipole radiators, realizing the transition from single-band to dual-band and increasing the gain. After loading the TDLS metasurface, the DP antenna, whose total height is 0.46λ0 and planar reflector at the bottom is 0.77λ0×0.77λ0 (where λ0 is the wavelength at resonance point 2.3 GHz), realizes two broad bands of 2.05-2.55 GHz (21.74%) and 3.09-3.71 GHz (18.24%), reaching average gains of 8.86 dBi and 7.40 dBi respectively, which is about 3 dBi higher than that without metasurface, and has a stable radiation pattern. As well as, the 3-dB half-power beamwidth (HPBW) is stabilized at 63.5°±6.6°, and the isolation is more splendid than 30 dB. Eventually, the radiation efficiency stabilized at around 85%.


  • Open Access

    Radar Spectrum Image Classification Based on Deep Learning

    by Zhongsen Sun, Kaizhuang Li, Yu Zheng, Xi Li, Yunlong Mao

    With the continuous development and progress of science and technology, the increasingly complex electromagnetic environment and the research and development of new radar systems have led to the emergence of various radar signals. Traditional methods of radar emitter identification cannot meet the needs of current practical applications. For the purpose of classification and recognition of radar emitter signals, this paper proposes an improved EfficientNetv2-s classification method based on deep learning for more precise classification and recognition of radar radiation source signals. Using 16 different types of radar signal parameters from the signal parameter setting table, the proposed method generates random data sets consisting of spectrum images with varying amplitude. The proposed method replaces two-dimensional convolution in EfficientNetV2 with one-dimensional convolution. Additionally, the channel attention mechanism of the EfficientNetv2-s is optimized and modified to obtain attention weights without dimensional reduction, resulting in superior accuracy. Compared with other deep-learning image-classification methods, the test results of this method have better classification accuracy on the test set: the top1 accuracy reaches 98.12%, which is 0.17~3.12% higher than other methods. Furthermore, the proposed method has lower complexity compared to most methods.