PWM-Based Impedance Boosting Technique With Autonomous Background Calibration for VCO-Based Neural Front Ends

Abstract

Adaptive impedance boosting for bio-signal acquisition front end is essential for wearable and implantable devices, where the sensor exhibits a high source impedance with a large spread. A pulsewidth-modulation (PWM)-based capacitively coupled chopped voltage-controlled oscillator (VCO)-based continuous-time $\mathrm {\Delta \Sigma }$ modulator (CTDSM) for bio-potential monitoring is presented in this article. A PWM-based positive feedback loop is proposed to cancel the coupling and decoupling effect inside the chip, thus boosting the input impedance with simplified calibration digital to analog converter (DAC). A frequency-domain loop stability detection is proposed to continuously monitor the VCO outputs with a fully digital implementation. Therefore, a background auto-calibration scheme is developed to achieve sub-second convergence time. The proposed VCO-based neural front end was fabricated in a 180-nm CMOS process. With a chopping frequency of 640 kHz, the prototype achieves 1.84- $\mu $ Vrms input-referred noise (IRN) from 0.07-Hz to 1-kHz bandwidth. With a linear input range of 150 mVpp, it exhibits an signal to noise and distortion ratio (SNDR) of 72.4 dB and a dynamic range (DR) of 90.2 dB. With a chopping frequency of 10 kHz, it exhibits an IRN of $3.09~{\mu }$ Vrms, an SNDR of 81.0 dB, and a DR of 85.7 dB. In addition, an input impedance of 8.73 G $\Omega $ is achieved at 2 Hz.