A 32-Channel Neural-Recording Chip Achieving 117dB Intrinsic-CMRR and 100dB PSRR by CM-Tracking-Dynamic-Power-Rail...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Electrical Engineering

Background:

• Multi-channel neural recording is crucial for monitoring brain activity across regions.
• In-vivo common-mode interference (CMI) severely degrades signal quality in implantable neural recording chips.
• Existing techniques for common-mode rejection ratio (CMRR) in analog front-ends (AFEs) have limitations in multi-channel systems.

Purpose of the Study:

• To develop a 32-channel neural recording chip with enhanced common-mode interference suppression.
• To overcome the limitations of traditional common-mode replication and other techniques in multi-channel neural recording.
• To achieve a high total common-mode rejection ratio (T-CMRR) for improved signal quality.

Main Methods:

• Fabrication of a 32-channel neural recording chip using a 65 nm CMOS process.
• Implementation of a CM-tracking-dynamic-power-rail (CM-TDPR) instrumentation amplifier (IA) for high-gain mode.
• Development of a CM-canceling-in-idle-phase (CM-CIP) technique for low-gain mode.
• In-vivo validation using Sprague-Dawley rats.

Main Results:

• The CM-TDPR IA achieved 50 GΩ CM input impedance, 117 dB intrinsic CMRR (I-CMRR), and 100 dB power supply rejection ratio (PSRR), yielding an 87 dB T-CMRR in high-gain mode.
• The CM-CIP technique increased I-CMRR to 102 dB and matched signal-reference input impedance, achieving a 95 dB T-CMRR in low-gain mode.
• In-vivo experiments confirmed the chip's effective CMRR performance.

Conclusions:

• The proposed 32-channel neural recording chip effectively suppresses CMI in both high-gain and low-gain modes.
• The novel CM-TDPR and CM-CIP techniques significantly enhance CMRR performance for implantable neural recording.
• This advancement promises improved signal quality for multi-channel neural monitoring applications.