Related Experiment Videos
A low-noise demultiplexing system for active multichannel microelectrode arrays
1Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor 48109.
IEEE Transactions on Bio-Medical Engineering
|January 1, 1991
Summary
This study presents a novel low-noise demultiplexing system for reconstructing neural signals from microelectrode arrays. The system achieves high fidelity signal recovery with minimal noise and crosstalk.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Signal Processing
Background:
- Multiplexed microelectrode arrays are crucial for recording neural activity.
- Efficient demultiplexing is needed to recover high-fidelity multichannel neural signals.
Purpose of the Study:
- To develop and characterize a low-noise demultiplexing system for neural signal reconstruction.
- To enable the recovery of low-amplitude neural signals from multiplexed data.
Main Methods:
- Design and implementation of a multichannel demultiplexing system.
- Characterization of system performance including gain, bandwidth, noise, and crosstalk.
- Testing with simulated and actual neural signals.
Main Results:
- Achieved a per-channel gain of 68 dB and a bandwidth of 100 Hz to 6 kHz.
- Demonstrated an equivalent noise level of 13 microV rms and interchannel crosstalk below 3%.
- Successfully reconstructed neural signals as low as 20 microV.
Conclusions:
- The developed system effectively reconstructs multichannel neural signals with high fidelity.
- The system's design is scalable to 40 channels, offering broad applicability.
- On-chip probe function control via power supply signaling is integrated.