Skip to content

DoRF++: Spherical Representation Learning over Doppler Radiance Fields for Robust Wi-Fi Sensing

Aug 2026 · arXiv.org · Vol abs/2608.08381 · 0 citations · 31 references
Computer Science Engineering

TL;DR

DoRF++ is introduced, a spherical representation-learning model that applies quadrature-aware spherical Transformers directly to the spherical field for activity classification, and noticeably outperforms state-of-the-art Wi-Fi-based HAR methods in cross-user generalization accuracy.

Abstract

Motivated by the IEEE 802.11bf effort to standardize advanced WLAN sensing, interest in Wi-Fi Channel State Information (CSI) for passive, device-free, and privacy-preserving activity and gesture recognition has grown rapidly. Recent studies have shown that Doppler velocity projections extracted from CSI, which directly reflect human-motion velocity, enable more robust human activity recognition (HAR) and stronger generalization across users and unseen conditions. Nevertheless, reliable generalization under real-world variability remains a major challenge that hinders the adoption of Wi-Fi sensing. To address this challenge, we build on Doppler Radiance Fields (DoRF), adapting the concept of neural radiance fields (NeRF) from computer vision to Wi-Fi sensing. DoRF treats the extracted Doppler velocity projections as sparse virtual-camera views of human motion. From these views it recovers a regularized latent 3-D motion descriptor whose projections along learned effective Doppler directions explain the CSI observations. The recovered motion is then re-projected onto an equiangular grid of directions on the unit sphere, yielding a spherical representation of the underlying activity. Because DoRF produces a signal defined on the sphere, this paper introduces DoRF++, a spherical representation-learning model that applies quadrature-aware spherical Transformers directly to the spherical field for activity classification. Experiments on our collected hand-gesture dataset show that DoRF++ noticeably outperforms state-of-the-art Wi-Fi-based HAR methods in cross-user generalization accuracy, especially for difficult gestures when only a single multi-antenna receiver access point (AP) is available.

View source

Similar papers

Preprint Nov 2026

From Activity Recognition to Shift Detection: Variational Wi-Fi Sensing with Evidential Uncertainty

Human motion perturbs wireless channels, making commodity Wi-Fi devices a practical platform for device-free activity recognition. Yet Channel State Information (CSI) also depends strongly on the person, receiver position, recording session, and surrounding environment, so a classifier can continue producing activity l...

Luca Cotti, Marco Cominelli, F. Gringoli et al. · 0 citations
Review Open access 2026

Physics-Informed Wi-Fi Channel State Information Sensing for Device-Free Human Activity Recognition: A Survey

Wi-Fichannel state information (CSI) enables non-intrusive device-free human activity recognition (DFHAR) on commodity wireless infrastructure, yet the electromagnetic sensitivity that resolves sub-millimeter motion makes trained models fragile to environment, hardware, and user change. Prior surveys treat cross-domain...

Zhi-Hao Zhao, Ya-Bing Chen, Chao Chen et al. · 0 citations
#computer vision Sep 2026

DiFF: Doppler-informed Flow Matching for Human Motion Flow

Perceiving human motion via privacy-preserving 4D millimeter-wave (mmWave) radar is critical for next-generation human-robot interaction (HRI), where point cloud scene flow serves as a foundational motion representation. Yet the extreme sparsity and noise of 4D radar point clouds make non-rigid motion flow estimation s...

Kai Wang, Ming-Le Zhao · 0 citations
Open access 2026

Toward an Explainable Machine Learning Method for Wi-Fi CSI-Based Human Activity Recognition

Human activity recognition (HAR) systems often involve vision-based sensors or wearable devices, which can be intrusive and raise privacy concerns. They also require dedicated hardware. This paper proposes a method for HAR based on Wi-Fi Channel State Information (CSI) data that moves toward explainable model decisions...

S. Poce, G. Varriano, C. Carissimo et al. · 0 citations
Conference Open access Oct 2026

Geometry Over Depth: Multi-Node WiFi CSI Fusion for Robust Human Activity Recognition

WiFi Channel State Information (CSI) enables contactless, device-free Human Activity Recognition (HAR), but a single transmitter-receiver link observes motion from one fixed geometry and is sensitive to subject position. The common response is to absorb this fragility within increasingly deep models, at the cost of dat...

Mojtaba Mansouri Nejad, Azin Moradbeikie, David Beserra et al. · 0 citations
#machine learning Preprint Oct 2026

UniCSI Towards a Universal Wi-Fi CSI Encoder for Ubiquitous Human Sensing

Wi-Fi sensing promises to turn the everyday wireless signals that already surround us into ubiquitous sensors for human sensing. However, a fundamental obstacle is that CSI is acquired under diverse device-specific configurations, including different subcarrier counts, bandwidths, and carrier bands. Consequently, the r...

Daniel Eckhoff, Hua Kang, Zhi-Tang Chen et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.