Weiming Li

Weiming Li

Auto-generated profile, based on publicly available information

Weiming Li

Beijing Key Laboratory of Millimeter-Wave and Terahertz Wave Technology, School of Integrated Circuits and Electronics, Beijing Institute of Technology, Beijing, China

Research Focus & Metrics

Academic Career Length

20Years (2007-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 October 2022 in Sensors
Verified
Journal Article
Open access
PDF file
MXMingxu Xiang
Mingxu Xiang
WRWu Ren
Wu Ren
WLWeiming Li
Weiming Li
ZXZhenghui Xue
Zhenghui Xue
XJXinyue Jiang
Xinyue Jiang
The method of using millimeter-wave radar sensors to detect human vital signs, namely respiration and heart rate, has received widespread attention in non-contact monitoring. These sensors are compact, lightweight, and able to sense and detect various scenarios. However, it still faces serious problems of noisy interference in hardware, which leads to a low signal-to-noise ratio (SNR). We used a frequency-modulated continuous wave (FMCW) radar sensor operating at 77 GHz in an office environment to extract the respiration and heart rate of a person accustomed to sitting in a chair. Indeed, the proposed signal processing includes novel impulse denoising operations and the spectral estimation decision method, which are unique in terms of noise reduction and accuracy improvement. In addition, the proposed method provides high-quality, repeatable respiration and heart rates with relative errors of 1.33% and 1.96% on average compared with the reference values measured by a reliable smart bracelet.

2 November 2020 in IEEE Antennas and Wireless Propagation Letters
Verified
Journal Article
Closed access
SFShutao Fang
Shutao Fang
WLWeiming Li
Weiming Li
ZXZhenghui Xue
Zhenghui Xue
WRWu Ren
Wu Ren
In this paper, we present a hybrid optimization method for the synthesis of distributed array consists of two identical sub-arrays, which is mainly used on aeroplanes as the frond-end of communication system, aiming to lower the grating lobe level as much as possible. The proposed method is on the basis of a combination of differential evolution optimization and weighted l_1 convex optimization. In the procedure of optimization, a hybrid method of optimization is proposed to optimize element positions and excitations to obtain the optimum performance, meanwhile, multiple constraints are adopted to satisfy the application requirements. Numerical comparisons between different values of parameters are presented to assess the effectiveness of the proposed optimization method, showing the best result of optimization possibly got with different parameters.

3 June 2022 in IEEE Antennas and Wireless Propagation Letters
Verified
Journal Article
Closed access
PWPeipei Wang
Peipei Wang
WRWu Ren
Wu Ren
WRWu Ren
Wu Ren
ZXZhenghui Xue
Zhenghui Xue
WLWei-Ming Li
Wei-Ming Li
A dual-band beam-scanning antenna by rotation of two transmintarrays (TAs) operating at 20 and 30 GHz has been presented in this letter. TA's dual-band unit has three metal layers and demonstrates almost completely independent performance at two frequency bands. Two designed TAs have units of different sizes to achieve dual-band with different linear phases. By rotating the two TAs synchronously around the axis orthogonal to the planes of the TAs, it enables two-dimensional (2-D) beam scanning with maximum scanning angles of ±40°. The achieved simulated peak gains are 19.9 dBi, 20.5 dBi and 22.5 dBi, 23.2 dBi at phi = 0°, phi = 90° at 20 and 30 GHz, respectively. The maximum beam scanning losses are 1.7 dB, 2.6 dB and 1.5 dB, 2.7 dB. Moreover, the two beams at the two frequencies can basically point to the same direction at the same time. To validate the design, a prototype is fabricated and tested. The measured results are in good agreement with the simulated results.

3 January 2025 in IEEE Antennas and Wireless Propagation Letters
Verified
Journal Article
Closed access
DDDajun Ding
Dajun Ding
WRWu Ren
Wu Ren
ZXZhenghui Xue
Zhenghui Xue
WLWeiming Li
Weiming Li
In this letter, a low-profile spin-decoupled folded reflectarray antenna (SDFRA) with dual-circularly-polarized beam control is proposed, which can realize spin angular momentum (SAM) decoupling. Impressively, the substrate used in this antenna is reduced to 3 layers. The proposed SDFRA consists of a receiver-transmitter metasurface (RTMS), a polarization conversion reflection metasurface (PCRM), and a magneto-electric (ME) dipole antenna. Specifically, the RTMS can reflect the x -polarized waves and transmit its orthogonal polarization waves. Meanwhile, the transmitted waves can be decomposed into left-hand circular polarization (LHCP) and right-hand circular polarization (RHCP) waves. To confirm the concept, a prototype is fabricated and measured. The results indicate that the 3-dB gain bandwidth of the LHCP beam and the RHCP beam are 20.89% and 21.3%, respectively. The maximum realized gain of the LHCP and RHCP beams are 20.55 dBic and 20.52 dBic, corresponding to aperture efficiencies of 19.06% and 19.19%. The experimental results are in good agreement with the simulated results. With the advantages of broadband, low profile, and low cost, this antenna provides an effective solution for single-band dual-circularly polarized beam control.

1 April 2012 in Acta Mechanica Sinica
Verified
Journal Article
Closed access
PDF file
WLWei-Ming Li
Wei-Ming Li
JHJia-Zhen Hong
Jia-Zhen Hong

1 July 2018
Verified
Conference Paper
Open access
ZWZairan Wang
Zairan Wang
WLWeiming Li
Weiming Li
YKYueying Kao
Yueying Kao
DZDongqing Zou
Dongqing Zou
QWQiang Wang
Qiang Wang
SHSung-Hoon Hong
Sung-Hoon Hong
Object pose estimation from a single image is a fundamental and challenging problem in computer vision and robotics. Generally, current methods treat pose estimation as a classification or a regression problem. However, regression based methods usually suffer from the issue of imbalanced training data, while classification methods are difficult to discriminate nearby poses. In this paper, a hybrid CNN model, which we call it HCR-Net that integrates both a classification network and a regression network, is proposed to deal with these issues. Our model is inspired by that regression methods can get better accuracy on homogeneously distributed datasets while classification methods are more effective for coarse quantization of the poses even if the dataset is not well balanced. The classification methods and the regression methods essentially complement each other. Thus we integrate both them into a neural network in a hybrid fashion and train it end-to-end with two novel loss functions. As a result, our method surpass the state-of-the-art methods, even with imbalanced training data and much less data augmentation. The experimental results on the challenging Pascal3D+ database demonstrate that our method outperforms the state-of-the-arts significantly, achieving improvements on ACC and AVP metrics up to 4% and 6%, respectively.