Sentinel is an open-source WiFi-Sensing framework designed to analyze physical movement and human presence within an environment by monitoring Radio Frequency (RF) disturbances in the 802.11 protocol. Unlike traditional surveillance systems, Sentinel prioritizes data privacy by processing only signal fluctuations (Jitter) and Amplitude variations (RSSI), rather than capturing visual or audio data.

The system is engineered across three primary functional layers:

1. Data Acquisition (Sniffer): The WiFi interface is configured in Monitor Mode to intercept packets (Beacon/Data) from a target BSSID. It extracts Raw RSSI data from the Radiotap headers.
2. Edge Processing (Backend): Captured telemetry is transmitted to an asynchronous FastAPI-based server. Signals are normalized and processed as time-series data to identify environmental anomalies.
3. Visualization (Frontend): A web-based interface utilizes Socket.io to render real-time data into a graphical radar format for end-user monitoring.

Sentinel is fully compatible with modern Linux distributions, including Fedora, Ubuntu, and Debian. The deployment process is streamlined via an automation script that manages system dependencies and kernel-level network configurations.

• WiFi network interface controller (NIC) with Monitor Mode support.
• Python 3.8 or higher.
• Root privileges (Sudo).

git clone [https://github.com/erogluyusuf/WiFi-Sensing-Radar.git](https://github.com/erogluyusuf/WiFi-Sensing-Radar.git)
cd WiFi-Sensing-Radar
sudo chmod +x install.sh
sudo ./install.sh

Sentinel utilizes the systemd service architecture for high availability and persistence. Upon installation, the following services are automatically configured to run as background processes:

To verify the operational status of the sensing engine, use the following command:

sudo systemctl status sentinel-sniffer

sudo ./uninstall.sh

Sentinel does not record imagery or audio. By analyzing only electromagnetic spectrum disturbances, it provides a security solution that adheres to the principles of "Privacy by Design." This methodology ensures the system is inherently compliant with global data protection standards, such as GDPR and KVKK, as it eliminates the collection of personally identifiable visual information.

Yusuf Eroğlu Computer Programmer & Management Information Systems Graduate

This project is licensed under the MIT License.

The MIT License is a permissive free software license, meaning anyone is free to fork, modify, and distribute this software for both personal and commercial use, provided the original copyright notice and permission notice are included. We encourage developers to contribute to the repository and enhance the sensing algorithms.

See the LICENSE file for further details.