Abstract
Radar-based human activity recognition (HAR) holds significant application value in fields such as medical rehabilitation and security monitoring. However, existing HAR methods primarily address line-of-sight (LOS) and non-LOS through-wall (NLOS-TW) scenarios, neglecting consideration for non-LOS corner (NLOS-C) scenario within urban architecture. In the NLOS-C scenario, electromagnetic waves illuminate the target through multiple paths, resulting in considerable variations in the range–time map, causing performance degradation or even failure of existing methods. Moreover, multipath propagation enables a single-node radar to function as a multiperspective multinode radar system, providing more comprehensive information for human activity. Therefore, considering the complementary interpretations of multipath and the distinctive features observed in the NLOS-C range–time map, this article proposes a HAR method for the NLOS-C scenario based on multipath fusion. First, considering the broad distribution and large span characteristics of behavior features in the range dimension caused by the multipath effect, we design a multipath information fusion module based on dilated convolution to effectively integrate and interact with the multipath information. In addition, to address the diverse feature scales caused by variable widths and blurred boundaries of each path, we incorporate a multiscale unit into the deep feature extraction module to enhance the capability of autonomously adjusting the receptive field. Finally, multipath interaction information is fused with depth features for recognition. Experimental results validate the effectiveness of the proposed method in the NLOS-C scenario. The codes are available at: https://github.com/tlz1111/Multipath-Fusion-Network
Funding Information
  • National Natural Science Foundation of China (62371110)
  • Municipal Government of Quzhou (2024D050)
  • Natural Science Foundation of Sichuan Province (2025ZNSFSC0467)