This repository contains the complete design files for the BioGAP v2, an integrated, ultra-low-power edge-AI platform for biosensing applications.
The BioGAP v2 brings together modular sensing shields, robust edge processing, and flexible debugging in a single, cohesive system. It includes the main processing board and a range of specialized shields for biopotential (ExG, EMG) and PPG sensing. A dedicated debug board and a reference template shield support rapid prototyping and customization for diverse biomedical and wearable monitoring applications.
-
Mainboard
High-performance core board with GAP9 and NRF5340 SoCs, optimized for edge AI and low-power wireless connectivity. -
ExG Shield
Flexible 16-channel biopotential sensing shield supporting both active and passive electrodes, fully differential or common reference configurations. -
EMG Shield
Dedicated EMG acquisition shield with 16 channels, swappable electrode-to-channel mapping PCBs, and balanced analog operation. -
PPG Shield
Compact two-channel PPG module (red & IR LED) for heart rate and SpO2 monitoring. Easily connects via cable for flexible placement. -
Debug Board
Simplifies flashing and debugging with dedicated debug connectors for all supported SoCs, signal breakouts, status LEDs, and user buttons. -
Template Shield
Reference PCB template with pre-routed connectors, serving as a starting point for custom sensing shield development.
- Modular stackable architecture for multi-modal physiological sensing.
- Ultra-low-power edge AI with GAP9’s parallel processing and NE16 accelerator.
- Wireless BLE connectivity via NRF5340.
- Versatile electrode configurations for a range of biosignal recordings.
- User-friendly debugging and expansion with standardized board-to-board connectors.
A detailed changelog is available in the Changelog.md file, documenting major updates and design revisions for each board.
- Sebastian Frey (sefrey@iis.ee.ethz.ch)
This repository makes use of the following licenses:
- for all hardware: Solderpad Hardware License Version 0.51
- for all images: Creative Commons Attribution 4.0 International License
For further information, please refer to the license files: LICENSE.hw
, LICENSE.images
.