Related Experiment Video
Updated: Jul 23, 2025
11:18
Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
Published on: March 2, 2015
10.4K
Design and Simulation of a Low Power 384-channel Actively Multiplexed Neural Interface
Gabriella Shull1, Yieljae Shin2, Jonathan Viventi1
1Department of Biomedical Engineering, Duke University, Durham, North Carolina.
Summary
This study introduces a novel 384-channel neural interface using CMOS technology and transfer printing. This high-density, actively multiplexed array significantly improves brain-computer interfaces (BCIs) by overcoming wiring limitations.
Area of Science:
- Neuroscience
- Electrical Engineering
- Biomedical Engineering
Background:
- Brain-computer interfaces (BCIs) restore lost functions but are limited by low resolution and high channel count wiring.
- Scaling neural interfaces is challenging due to output wiring and connector size constraints.
Purpose of the Study:
- To design and simulate a high-density, actively multiplexed neural recording array.
- To overcome the limitations of existing BCIs in terms of cortical coverage and resolution.
Main Methods:
- Leveraged a 130-nm CMOS process and transfer printing for array fabrication.
- Designed a 384-channel actively multiplexed array with front-end filtering and amplification at each electrode site.
- Utilized time division multiplexing (TDM) to reduce wiring complexity.
Main Results:
- Achieved 50 μm × 50 μm pixels enabling recording of 384 channels at 30 kHz.
- Demonstrated a low noise level of 9.57 μV rms with a gain of 22.3 dB.
- Reported a low power consumption of 0.63 μW/channel with a wide bandwidth (0.1 Hz - 10 kHz).
Conclusions:
- The developed actively multiplexed array minimizes noise and overcomes wiring limitations for neural interfaces.
- This technology enables high channel-count arrays, paving the way for improved BCIs.
- The design is broadly applicable to various neural interface applications requiring high-density recordings.
