This repository is a curated list of research papers, datasets, and tools focused on WiFi Channel State Information (CSI) sensing. It serves as a comprehensive resource for researchers and practitioners interested in leveraging WiFi signals for various sensing applications, such as human activity recognition, occupancy detection, and indoor localization.

How It Works

The project categorizes resources by application area, method, and platform. It highlights papers employing learning-based, modeling-based, and hybrid approaches to interpret CSI data. The collection also details various CSI extraction tools (e.g., Intel 5300, Atheros, Nexmon) and significant datasets, providing a structured overview of the WiFi CSI sensing landscape.

Quick Start & Requirements

This is a curated list, not a software package. To utilize the resources:

• CSI Tools: Installation varies by tool (e.g., Intel 5300 CSI Tool, Atheros CSI Tool, Nexmon CSI Tool). Specific instructions are typically found within each tool's repository.
• Libraries: Many associated code repositories are linked, often requiring Python and deep learning frameworks like PyTorch or TensorFlow.
• Datasets: Downloading and processing datasets will require significant storage and computational resources.

Highlighted Details

• Extensive categorization of applications including Occupancy Detection, Human Activity Recognition, Human Identification, Crowd Counting, Gesture Recognition, Fall Detection, Vital Sign Detection, In-Car Activity Recognition, Pose Estimation, and Indoor Localization.
• Covers emerging areas like LLM+WiFi-CSI Sensing and challenges in large-scale, cross-environment, and multi-modal sensing.
• Lists multiple CSI extraction platforms (Intel 5300, Atheros, Nexmon, ESP32, SDR) and key datasets (MM-Fi, NTU-Fi, Widar 3.0).
• Provides links to numerous associated GitHub repositories for code implementations of listed papers.

Maintenance & Community

The repository is community-driven, welcoming suggestions via pull requests. It cites the "SenseFi: A Library and Benchmark on Deep-Learning-Empowered WiFi Human Sensing" paper as a foundational work.

Licensing & Compatibility

The repository itself is a list and does not have a specific license. Individual linked papers and code repositories will have their own licenses, which must be consulted for usage and compatibility, especially for commercial applications.

Limitations & Caveats

This is a reference list, not an executable framework. Users must independently acquire, install, and configure the various CSI tools, libraries, and datasets mentioned. The rapid evolution of the field means some resources may become outdated.