batear transforms an ESP32-S3 microcontroller into an affordable drone detection system for under $15. This edge-only solution detects drone rotor harmonics using a MEMS microphone, requiring no internet access or cloud subscriptions. Its continuous monitoring and low power consumption make it ideal for any home environment.
batear: An Affordable Acoustic Drone Detection System
batear is an innovative solution designed to protect homes, farms, and communities from the increasing threat posed by drones. By leveraging a low-cost hardware setup of under $15, this project transforms the compact ESP32-S3 microcontroller in conjunction with a MEMS microphone into an effective and autonomous drone detection system.
This edge-only detection technology operates without any need for cloud services, internet connection, or ongoing costs, ensuring both privacy and affordability. Simply deploy the device at strategic locations such as windows, rooftops, or fences, and it will promptly alert when drone rotor harmonics are detected in the vicinity.
Key Technical Features
The batear system utilizes advanced audio processing through an ICS-43434 I2S MEMS microphone. It implements multi-frequency Goertzel filters to analyze tonal energy corresponding to drone rotor sounds. An alarm is triggered when the tonal to broadband energy ratio surpasses a predetermined threshold. The Goertzel algorithm runs efficiently within the ESP32-S3's 512 KB SRAM while consuming minimal power, making it suitable for battery-operated or solar-powered implementations.
"Built for defense, hoping it becomes unnecessary. A vision where no one needs to fear the sky."
Important Considerations
While batear offers a promising solution, practical deployment depends on various environmental factors such as:
- Distance from drones
- Wind Conditions
- Background Noise levels
- Types of Drones
Due to these variables, calibration of thresholds specific to the environment is critical for achieving optimal accuracy. The project serves as a flashable baseline, with enhanced accuracy possible through ESP-NN or TensorFlow Lite Micro models.
Hardware Connection Diagram
Connectivity with the ICS-43434 MEMS microphone is straightforward:
| ICS-43434 | T-Display-S3 |
|---|---|
| VDD | 3.3V |
| GND | GND |
| SCK | GPIO43 (BCLK) |
| WS | GPIO44 (LRCLK / WS) |
| SD | GPIO1 (DIN) |
| L/R | GND (left channel) |
Configuration Parameters
Key adjustable parameters for effective performance include:
| Symbol | Default | Description |
|---|---|---|
SAMPLE_RATE_HZ | 16000 | Sampling frequency |
FRAME_SAMPLES | 512 | Number of samples analyzed per frame |
HOP_MS | 100 | Time interval between analyses (ms) |
FREQ_RATIO_ON | 0.008 | Threshold to trigger alarm |
FREQ_RATIO_OFF | 0.004 | Threshold to clear alarm |
Additionally, the project is currently in the Hardware Proof-of-Concept (PoC) phase. The DSP logic has been verified in controlled indoor settings, effectively identifying rotor harmonics and operating within the ESP32's constraints. However, addressing challenges such as wind noise and mechanical interference in outdoor environments is part of future development.
Future Prospects
The next steps involve collecting real-world outdoor data to refine detection algorithms, with the possibility of training a lightweight neural network model for enhanced performance.
The project structure is neatly organized, allowing for easy navigation and management.
For anyone interested in implementing an effective and low-cost drone detection solution, batear offers a compelling approach, integrating advanced technology with practicality and affordability.
No comments yet.
Sign in to be the first to comment.