Physics-Guided Attention in a Lightweight TCN for Efficient WiFi CSI-Based Human Activity Recognition

2026-06-01

AI Summary

This work addresses the high computational cost and inefficiency of existing WiFi channel state information (CSI)-based human activity recognition methods, which rely on complex deep models to implicitly learn motion dynamics. To overcome these limitations, the authors propose a lightweight temporal convolutional network (TCN) that explicitly models physical motion characteristics in CSI signals by integrating Doppler energy–guided temporal attention with variance-driven subcarrier channel attention. This design embeds domain-specific priors without increasing network depth. Evaluated on multiple benchmark datasets, the proposed method significantly outperforms existing deep learning baselines in both recognition accuracy and inference efficiency, achieving superior performance with fewer parameters and lower computational overhead.

Abstract

Human Action Recognition (HAR) using WiFi Channel State Information (CSI) has gained increasing attention due to its non-contact, low-cost, and privacy-preserving nature. However, existing learning-based approaches largely rely on deep, computationally intensive architectures to implicitly capture motion dynamics from CSI measurements, thereby increasing model complexity and reducing efficiency. Instead, we argue that incorporating appropriate inductive biases tailored to the physical characteristics of CSI signals enables more efficient and effective learning. In this work, we propose a compact temporal convolutional network (TCN)-based framework that explicitly incorporates motion-aware inductive biases into feature learning. Specifically, we introduce a Doppler-energy-guided temporal attention mechanism in feature space to emphasize motion-salient time segments, and a variance-driven channel attention module to weight informative subcarriers based on temporal motion statistics adaptively. By integrating these domain-specific priors, the proposed model effectively captures motion dynamics without increasing architectural depth. Extensive experiments on multiple benchmark datasets demonstrate that our approach achieves superior performance compared to deeper baselines, while significantly reducing parameter count and computational cost.

Problem

• WiFi CSI
• Human Activity Recognition
• Model Efficiency
• Motion Dynamics
• Computational Complexity

Innovation

• Physics-Guided Attention
• Lightweight TCN
• Doppler-Energy Guidance
• Variance-Driven Channel Attention
• CSI-Based HAR