Skip to content

ed0ted/microphone-location-v1

BranchesTags

Repository files navigation

Acoustic Drone Localization System

A real-time 3D acoustic localization system for drone detection using distributed microphone arrays on Raspberry Pi nodes.

🎯 Features

  • 3-Microphone Triangle Arrays - Optimized horizontal arrays on each node
  • Real-time 3D Tracking - Interactive web-based visualization
  • Distributed Architecture - 3× Pi Zero 2 W nodes + 1× Pi 5 fusion server
  • WiFi Access Point Mode - Low-latency local network
  • Flexible Deployment - Supports both triangle (3-mic) and tetrahedron (4-mic) modes
  • WSL Testing - Full simulation mode for development without hardware

📋 Hardware Requirements

Per Node (×3):

  • Raspberry Pi Zero 2 W
  • ADS1115 16-bit ADC (I²C)
  • 3× GY-MAX4466 microphone modules
  • Power supply (5V, 2A)

Central Server:

  • Raspberry Pi 5 (4GB+ RAM)
  • Display (HDMI)
  • Power supply (27W USB-C)

Total Cost: ~$300-400 for complete 3-node system

🚀 Quick Start

Option 1: Automated Setup (Recommended)

On each Pi Zero node:

cd /path/to/microphone-location-v2
sudo ./scripts/setup-node.sh 1  # Use 1, 2, or 3 for node ID

On Pi 5 server:

cd /path/to/microphone-location-v2
sudo ./scripts/setup-server.sh

Option 2: Manual Setup

Node Setup:

sudo apt update && sudo apt install -y python3-venv python3-pip i2c-tools
python3 -m venv /opt/drone-node
source /opt/drone-node/bin/activate
pip install -r requirements-node.txt

# Configure and calibrate
nano configs/node-1.yaml
python -m node.node_agent --config configs/node-1.yaml calibrate --duration 60
python -m node.node_agent --config configs/node-1.yaml run

Server Setup:

sudo apt update && sudo apt install -y python3-venv python3-pip
python3 -m venv /opt/drone-server
source /opt/drone-server/bin/activate
pip install -r requirements-server.txt

python -m server.run_all --config configs/server.yaml

Access Web Interface: http://<pi5-ip>:80

🧪 Testing Without Hardware (WSL)

Full simulation mode for development on Windows:

# In WSL Ubuntu
cd ~/microphone-location-v2

# Update configs for localhost
sed -i 's/192.168.50.1/127.0.0.1/g' configs/*.yaml

# Terminal 1: Server
python -m server.run_all --config configs/server.yaml --verbose

# Terminals 2-4: Nodes
python -m node.node_agent --config configs/node-1.yaml run
python -m node.node_agent --config configs/node-2.yaml run
python -m node.node_agent --config configs/node-3.yaml run

Web Interface: http://localhost:8080

See WSL Testing Guide for detailed instructions.

📖 Documentation

Setup Guides

Technical Documentation

🏗️ System Architecture

┌─────────────────────────────────────────────────────┐
│                Surveillance Area                     │
│                   (~20m × 20m)                       │
│                                                      │
│    Node 1              Drone ✈️           Node 3    │
│     🎤🎤🎤                                 🎤🎤🎤     │
│    (0,0,1)                               (0,20,1)   │
│                                                      │
│                   Node 2                             │
│                    🎤🎤🎤                              │
│                   (20,0,1)                           │
└──────────────────────┬──────────────────────────────┘
                       │ UDP over WiFi
                       ↓
              ┌─────────────────┐
              │  Raspberry Pi 5  │
              │  Fusion Server   │
              │   192.168.50.1   │
              └────────┬─────────┘
                       │
                   📺 Display
              3D Position Tracking

Data Flow:

  1. Each node samples 3 microphones at ~860 Hz
  2. Node computes RMS energy, direction vector, noise floor
  3. Node sends UDP packet every ~100ms to Pi 5
  4. Pi 5 fuses data from all 3 nodes
  5. Pi 5 estimates 3D drone position via grid search
  6. Web UI displays position in real-time

📁 Repository Structure

microphone-location-v2/
├── node/                      # Pi Zero node package
│   ├── node_agent.py         # Main node agent
│   ├── ads_sampler.py        # ADC sampling (real + simulated)
│   ├── dsp.py                # Signal processing & feature extraction
│   ├── config.py             # Configuration management
│   └── packets.py            # Network packet handling
│
├── server/                    # Pi 5 fusion server
│   ├── run_all.py            # Main entry point
│   ├── localization.py       # 3D localization algorithm
│   ├── fusion_receiver.py    # UDP packet receiver
│   ├── state_store.py        # Shared state management
│   ├── config.py             # Server configuration
│   └── web/                  # Web interface
│       ├── app.py            # Flask + Socket.IO server
│       ├── templates/        # HTML templates
│       │   └── index.html
│       └── static/           # JS, CSS assets
│           ├── main.js       # 3D visualization (Three.js)
│           └── styles.css    # Modern UI styling
│
├── configs/                   # Configuration files
│   ├── node-1.yaml           # Node 1 config
│   ├── node-2.yaml           # Node 2 config
│   ├── node-3.yaml           # Node 3 config
│   └── server.yaml           # Fusion server config
│
├── scripts/                   # Deployment scripts
│   ├── setup-node.sh         # Automated node setup
│   ├── setup-server.sh       # Automated server setup
│   ├── launch-node.sh        # Node launcher
│   ├── launch-server.sh      # Server launcher
│   └── calibrate-node.sh     # Calibration helper
│
├── docs/                      # Documentation
│   ├── hardware-setup.md     # Hardware assembly guide
│   ├── wifi-access-point.md  # WiFi AP configuration
│   ├── wsl-testing.md        # WSL testing guide
│   └── implementation-plan.md # Technical design doc
│
├── requirements-node.txt      # Node Python dependencies
├── requirements-server.txt    # Server Python dependencies
└── README.md                  # This file

🔧 Configuration

Triangle Array Mode (3 Microphones)

Each node uses 3 microphones in a horizontal equilateral triangle, all pointing upward:

array_mode: triangle

triangle_positions:
  - [0.15, 0.0, 0.0]      # Mic 0: Right
  - [-0.075, 0.13, 0.0]   # Mic 1: Back-Left
  - [-0.075, -0.13, 0.0]  # Mic 2: Front-Left

triangle_vectors:
  - [0.0, 0.0, 1.0]       # All point directly upward
  - [0.0, 0.0, 1.0]       # Optimal for elevated sources (drones)
  - [0.0, 0.0, 1.0]       # Horizontal direction from spatial diversity

Tetrahedron Mode (4 Microphones)

For improved elevation sensing, switch to 4-mic tetrahedron mode:

array_mode: tetrahedron

tetrahedron_positions:
  - [0.0, 0.0, 0.18]      # Top
  - [0.0, 0.17, -0.06]    # Front
  - [-0.15, -0.09, -0.06] # Back-Left
  - [0.15, -0.09, -0.06]  # Back-Right

🎛️ Usage

Running the System

Start all services automatically:

On each node:

sudo systemctl enable drone-node
sudo systemctl start drone-node

On server:

sudo systemctl enable drone-server
sudo systemctl start drone-server

Or run manually:

# On nodes
./scripts/launch-node.sh 1  # Node 1
./scripts/launch-node.sh 2  # Node 2
./scripts/launch-node.sh 3  # Node 3

# On server
./scripts/launch-server.sh

Calibration

Before first use, calibrate ambient noise levels:

./scripts/calibrate-node.sh 1 60  # Node 1, 60 seconds

Keep environment QUIET during calibration!

Monitoring

Check system status:

sudo systemctl status drone-node   # On nodes
sudo systemctl status drone-server # On server

View logs:

sudo journalctl -u drone-node -f   # Node logs
sudo journalctl -u drone-server -f # Server logs

Monitor network:

# On server
sudo tcpdump -i wlan0 udp port 5005

📊 Performance

Expected Accuracy

With this hardware configuration:

Metric Best Case Typical Notes
Horizontal (X-Y) 3-5 m 5-8 m Depends on SNR
Vertical (Z) 4-6 m 6-10 m Limited with triangle mode
Detection Range 30 m 20 m Clear line-of-sight
Update Rate 10 Hz 5 Hz Network dependent
Latency 100 ms 200 ms WiFi AP mode

Limitations

  • Coarse localization (meters, not centimeters) due to MAX4466/ADS1115 sampling rate
  • SNR dependent - accuracy degrades with distance and ambient noise
  • No TDoA - uses intensity-based localization only
  • Triangle mode has reduced elevation sensing vs tetrahedron

🐛 Troubleshooting

Problem: Node won't detect ADS1115

# Check I²C
sudo i2cdetect -y 1
# Should see 0x48

# Enable I²C
sudo raspi-config  # Interface Options -> I2C -> Enable

Problem: No data reaching server

# Check network
ping 192.168.50.1  # From node

# Check firewall
sudo ufw allow 5005/udp

# Check server logs
sudo journalctl -u drone-server -f

Problem: Poor localization

  • Calibrate nodes in quiet environment
  • Check node positions in configs/server.yaml
  • Increase SNR - fly drone closer
  • Adjust grid search - smaller grid_step (slower but more accurate)

See Hardware Setup for more solutions.

🔬 Development

Running Tests

pytest tests/ -v
pytest tests/ --cov=node --cov=server

Code Style

black node/ server/
flake8 node/ server/
mypy node/ server/

Adding Features

  1. Fork the repository
  2. Create feature branch: git checkout -b feature/my-feature
  3. Test thoroughly in simulation mode
  4. Test on real hardware
  5. Submit pull request

📝 License

[Add your license here]

🙏 Acknowledgments

  • Based on acoustic drone detection research
  • Uses Three.js for 3D visualization
  • Plotly for real-time charts
  • Flask + Socket.IO for web interface

📧 Contact

[Add contact information]

Version: 2.0
Last Updated: November 2024
Hardware: Raspberry Pi Zero 2 W, Raspberry Pi 5, ADS1115, MAX4466

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages