Millimeter-Wave Radar-Based Multi-Human Tracking and Fall Detection System

Overview

This repository contains the implementation and code resources for our paper: Advanced Millimeter-Wave Radar System for Real-Time Multiple-Human Tracking and Fall Detection (link: https://doi.org/10.3390/s24113660). The study explores an indoor system that employs Millimeter-Wave radars to track multiple humans and detect falls in real time. By integrating signals from non-intrusive radars, our framework addresses challenges such as mobility inconvenience, lighting conditions, and privacy issues inherent in wearable or camera-based systems.

Key Features

• Multi-Human Tracking: Tracks multiple humans simultaneously with high precision.
• Real-Time Fall Detection: Accurately predicts and classifies human body statuses, including falls.
• Advanced Signal Processing: Employs Dynamic DBSCAN clustering and innovative feedback loops for enhanced accuracy.
• Privacy and Accessibility: Operates without cameras or wearables, ensuring non-intrusive monitoring. Camera module in the project is just for ground truth.

---

Table of Contents

1. System Architecture
2. Installation and Usage
3. License

---

System Architecture

Components

• Radar Hardware: 3 Millimeter-Wave radars from Texas Instruments.
• Real-Time Framework: Integrates radar signals to track and classify human activity.

Workflow

---

Installation and Usage

Prerequisites

• Python 3.8 or higher
• Libraries: numpy, Send2Trash,scipy,pyserial,matplotlib,scikit-learn,opencv-python,google-api-python-client,google-auth-oauthlib,func-timeout,moviepy

Steps

1. Clone this repository:

   git clone https://github.com/DarkSZChao/MMWave_Radar_Human_Tracking_and_Fall_detection.git
   cd MMWave_Radar_Human_Tracking_and_Fall_detection
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Connect radars.

4. Check which port number the radars are using:

5. Config the parameters:

   cd cfg

   open config_demo.py, under RADAR_CFG_LIST parameter, update cfg_port_name and data_port_name for radars

6. Go back to root folder and start the system by runing main.py

---

License

This project is licensed under the MIT License.

You are free to use, modify, and distribute this project, provided that you include the original copyright and license notice in any copy of the project or substantial portions of it.

See the LICENSE file for more details.