Beyond Amplitude: Channel State Information Phase-Aware Deep Fusion for Robotic Activity Recognition

TL;DR

This paper introduces GF-BiLSTM, a novel deep learning model that effectively fuses Wi-Fi CSI amplitude and phase information for improved robotic activity recognition, demonstrating the importance of phase data.

Contribution

It is the first systematic exploration of CSI phase in robotic activity recognition, showing how phase enhances recognition accuracy and robustness.

Findings

Incorporating phase data improves recognition accuracy.

GF-BiLSTM outperforms other models in experiments.

Phase information enhances cross-speed robustness.

Abstract

Wi-Fi Channel State Information (CSI) has emerged as a promising non-line-of-sight sensing modality for human and robotic activity recognition. However, prior work has predominantly relied on CSI amplitude while underutilizing phase information, particularly in robotic arm activity recognition. In this paper, we present GateFusion-Bidirectional Long Short-Term Memory network (GF-BiLSTM) for WiFi sensing in robotic activity recognition. GF-BiLSTM is a two-stream gated fusion network that encodes amplitude and phase separately and adaptively integrates per-time features through a learned gating mechanism. We systematically evaluate state-of-the-art deep learning models under a Leave-One-Velocity-Out (LOVO) protocol across four input configurations: amplitude only, phase only, amplitude + unwrapped phase, and amplitude + sanitized phase. Experimental results demonstrate that incorporating phase alongside amplitude consistently improves recognition accuracy and cross-speed robustness, with GF-BiLSTM achieving the best performance. To the best of our knowledge, this work provides the first systematic exploration of CSI phase for robotic activity recognition, establishing its critical role in Wi-Fi-based sensing.