Wireless signal imaging technology

  • Human pose imaging technology based on commercial Wi-Fi equipment

In this study, we designed the multipath resolution improvement mechanism of Wi-Fi signal, studied the retention algorithm of interested components of Wi-Fi signal, constructed a cross-morphological supervision human pose imaging model, and enhanced the generalization ability of the model through domain antagonism design. Compared to most Wi-Fi action recognition technologies, they are more granular and have visual advantages.

We built the first Wi-Fi human pose imaging system and designed an experimental test model. The experimental results show that commercial Wi-Fi devices can achieve human pose imaging accuracy similar to that of cameras. Moreover, commercial Wi-Fi devices can also achieve accurate imaging through walls.

  • Results
  • Papers

[1] Lingchao Guo, Zhaoming Lu, Xiangming Wen, Shuang Zhou, Zijun Han. (2019). From signal to image: Capturing fine-grained human poses with commodity Wi-Fi. IEEE Communications Letters24(4), 802-806.

[2] Shuang Zhou, Lingchao Guo, Zhaoming Lu, Xiangming Wen, Wei Zheng, Yiming Wang. Subject-independent Human Pose Image Construction with Commodity Wi-Fi. 2021 International Conference on Communications (ICC), (under review )

[3] Lingchao Guo, Xiangming Wen, Zhaoming Lu, Xinbin Shen, Zijun Han. (2019, April). WiRoI: Spatial region of interest human sensing with commodity WiFi. In 2019 IEEE Wireless Communications and Networking Conference (WCNC) (pp. 1-6). IEEE.

[4] Lingchao Guo, Zhaoming Lu, Shuang Zhou, Xiangming Wen, Zhihong He. (2020). When Healthcare Meets Off-the-Shelf WiFi: A Non-Wearable and Low-Costs Approach for In-Home MonitoringarXiv preprint arXiv:2009.09715.

  • 3D mobile human sensing technology based on Wi-Fi

3D human pose estimation can solve the problems such as limb loss and incomplete imaging in 2D human pose imaging. We built a prototype system for human pose estimation .The experimental results showed that the average positioning errors are 29.7mm and 37.8mm in LoS and NLoS scenarios respectively.

In addition, for the actual application scenarios such as monitoring of the elderly living alone and security monitoring, the perception of the moving human body not only needs to obtain the posture of the human body, but also the position information of the human body. The combination of human trajectory tracking and three-dimensional human posture estimation can provide a full range of three-dimensional moving human body perception. Therefore, we also built a track tracking system, which can realize decimeter-level human tracking in the 36.0 m^2 perception area.

  • Results
  • Papers

[1] Yiming Wang, Lingchao Guo, Zhaoming Lu, Xiangming Wen, Shuang Zhou, Wanyu Meng. (2021). From Point to Space: 3D Moving Human Pose Estimation Using Commodity WiFi. IEEE Communications Letters.

  • Electromagnetic wave imaging technique

We proposes a new indoor Wi-Fi imaging methods based on the theory of microwave imaging, the experiment only need to use ordinary commercial Wi-Fi routing equipment. CSI is utilized to extract the signal phase and amplitude of the object signal propagation in the space formation of the two-dimensional wave front front after the theory of scalar diffraction reconstruction to restore the three-dimensional view of objects.

At present,the renderings of indoor actual metal imaging have been realized, and patent pending trial proposal.

  •  Results