A 4.0-μVrms 7.5-kHz-BW Neural Signal Recording Circuit for Closed-Loop Deep Brain Stimulation

This work presents a low-noise biosignal acquisition application-specific integrated circuit (ASIC) for neural signal recording in closed-loop deep brain stimulation (DBS). An ac-coupled low-noise amplifier (LNA) amplifies the signals of interest while removing the low-frequency electrode offset. An improved programmable-gain amplifier (PGA) employs a flip-over-capacitor structure to prevent the degeneration of frequency response from switches’ off-state resistance. The differential-difference amplifier (DDA) used in the PGA isolates the switching noise from the following switched-capacitor low-pass filter that defines the upper limit of the bandwidth. A 12-bit analog-to-digital converter (ADC) converts the signals to digital representations with a sampling rate of 21.5 kS/s. The proposed circuit was fabricated in a 180 nm CMOS process with an area of 0.576 mm
²
, consuming 60 μA under a 1.8-V supply. It achieves a programmable gain from 47.3 dB to 59.1 dB. The measured input-referred noise is 1.74 μV
rms
(1 Hz–300 Hz) and 3.73 μV
rms
(300 Hz–7.5 kHz). The ADC used for digitization achieves an ENOB of 10.63 bits.