LWDSC-Net: A Lightweight Wavelet Dual-Branch Coupling Network for mmWave Radar-Based Human Activity Recognition

Previous studies on human activity recognition (HAR) using mmWave radar in IoT scenarios have relied on complex network architectures and heavy preprocessing, making them computationally expensive and difficult to deploy on resource-constrained edge devices (e.g., smart homes, healthcare). To address this, we propose LWDSC-Net, a lightweight wavelet dual-branch coupling network that embodies a novel, physically-interpretable design paradigm. Inspired by classic signal processing, LWDSC-Net embeds a 1D Haar wavelet transform (WT) as a front-end layer, enabling an end-to-end, structured decoupling of raw signals to construct a dual-branch structure with low and high frequencies (LF & HF) without external manual preprocessing. This signal-architecture co-design dedicates each branch to its respective frequency components, with the multi-level WT concept further extended to the HF branch. For the LF component branch, we propose an improved lightweight temporal convolutional network (LightTCN), which builds the feed-forward network using depthwise convolution. LightTCN reduces the depth of module stacking and efficiently models long-term global dependencies across the entire action sequence. For the HF components branch, we propose the wavelet-domain depthwise separable convolution (WD-DSC) module. It enhances micro-motion features by combining secondary wavelet reconstruction with wavelet-domain convolution, and further refines local detail information through deep convolutional layers. Experimental results show that LWDSC-Net achieves an accuracy of 98.76% on the public dataset while significantly reducing the number of model parameters and computational complexity. More importantly, deployment on an NVIDIA Jetson Nano confirms its excellent real-world performance, featuring ultra-low latency and power consumption, which highlights its outstanding potential for edge deployment.