摘要： 在人体行为识别领域中，毫米波雷达兼具复杂环境适应性与隐私保护优势，但现有基于毫米波雷达人体行为识别方法存在准确率低、数据表征不足、时间序列依赖关系捕捉难、计算资源消耗大的问题。为此，本文提出一种基于时序捕捉和增强模块（TCM-TMEM）的轻量化毫米波雷达人体行为识别方法。首先，设计时序捕捉模块（TCM），以因果卷积为基础提升局部时序敏感性并通过简化局部特征提取网络结构以降低单模块计算开销。其次，基于Transformer编码器设计时序增强模块（TMEM），利用其全局建模能力，强化网络对全局时序关联的捕捉能力，同时通过模块参数精简设计保障轻量化特性。然后，针对毫米波雷达距离-多普勒图表征能力不足，创新性引入包含距离、多普勒频移、信号能量等11个关键维度的特征构建方案，弥补传统数据维度表征信息不足的缺陷，提升了数据表征完整性。最后，在自建数据集PACT和公开数据集R-IHB上开展实验验证，结果显示该方法识别准确率达89.86%和86.63%，值得注意的是，TCM-TMEM模型仅0.12M，充分证明所提特征构建方案与模型在提升识别准确率、解决时序建模困难、降低计算资源消耗上的有效性。

Abstract: Millimeter-wave radar offers distinct advantages for human activity recognition, including robustness in complex environments and inherent privacy preservation. However, existing recognition methods confront several challenges: low accuracy, insufficient data representation, difficulty in modeling temporal dependencies, and high computational costs. To address these issues, this paper proposes a lightweight human activity recognition method based on a novel Time-series Capture and Enhancement Module (TCM-TMEM).The proposed architecture comprises two primary components. First, the Temporal Capture Module (TCM) employs causal convolution to enhance sensitivity to local temporal patterns, while its simplified network design minimizes computational overhead. Second, the Temporal Enhancement Module (TMEM) is constructed using a parameter-efficient Transformer encoder. This module strengthens the network's ability to model long-range, global temporal correlations while preserving the model's lightweight characteristics. Furthermore, to mitigate the representational limitations of traditional range-Doppler maps, an enhanced 11-dimensional feature set is introduced. This set incorporates critical dimensions such as range, Doppler shift, and signal energy, thereby significantly improving the completeness of data representation. Experimental evaluations were conducted on the self-collected PACT dataset and the public R-IHB dataset. The proposed method achieved recognition accuracies of 89.86% and 86.63%, respectively. Importantly, the entire TCM-TMEM model contains only 0.12 million parameters. These results substantiate the effectiveness of the proposed feature construction scheme and model architecture in improving recognition accuracy, effectively capturing temporal dependencies, and substantially reducing computational resource consumption.