This article presents a high-dynamic-range (DR) voltage-controlled oscillator (VCO)-based front end for sensor readout applications. Unlike conventional VCO-based quantizers, which suffer from large voltage-to-frequency non-linearities, the proposed design leverages differential pulse-code modulation (DPCM) from compression theory to substantially reduce the amplitude of the signal incident to the VCO quantizer, thereby achieving an ultra-low total harmonic distortion (THD) of -112 dB. In addition, background digital gain calibration is employed to overcome gain deviation of the VCO, thus ensuring a robust design. Together with dynamic element matching (DEM), the techniques enable a high DR using only the first-order noise shaping inherent in VCO-based quantizers and a moderate 32x oversampling ratio. Fabricated in 65 nm, the sensor front end consumes 3.2-μW power and achieves an SNDR of 89 dB and a DR of 94 dB in 500 Hz of bandwidth. Together with a 1.18-μVrms integrated input-referred noise, it achieves a noise efficiency factor (NEF) of 4 and a Schreier FoM of 171 dB.
23 January 2019 in IEEE Solid-State Circuits Letters
This letter presents a miniaturized neural recording analog front-end (AFE) suitable for distributed sub-mm neural recording implants. Including reference generators, this AFE occupies an area of only 0.01 mm2, thanks to a mostly digital architecture with more than half of area being scalable digital logic. Implemented in 65 nm CMOS, the AFE is measured to consume 3.2 μW of power from 0.6 V supply, and achieves an input-referred noise of 2.2 μVrms over 500 Hz of bandwidth. The AFE is validated by integrating it into a neural sensing SoC with wireless power and data telemetry. In vitro experiments demonstrate reliable recording of the AFE in a fully wireless environment.
7 December 2021 in IEEE Transactions on Biomedical Circuits and Systems
This paper presents a second-order voltage-controlled oscillator (VCO)-based front-end for the direct digitization of biopotential signals. This work addresses the non-linearity of VCO-based ADC architectures with a mismatch resilient, multi-phase quantizer, a gated-inverted-ring oscillator (GIRO), achieving >110-dB SFDR. Leveraging the time-domain encoding of the first integrator, the ADC's power is dynamically scaled with the input amplitude enabling up to 35% power savings in the absence of motion artifacts or interference. An auxiliary input-impedance booster increases the ADC's input impedance to 50 MΩ across the entire bandwidth. Fabricated in a 65-nm CMOS process, this ADC achieves 92.3-dB SNDR in a 1 kHz BW while consuming 5.8 µW for a 174.7 dB Schreier FoM.
23 September 2021 in IEEE Journal of Solid-State Circuits
This article presents a voltage-controlled oscillator (VCO)-based analog front end (AFE) for ExG readout applications with both a wide dynamic range (DR) and high linearity. By using a differential pulse code modulation (DPCM) technique, VCO non-linearity is mitigated by operating the VCO in the small-signal linear regime. To minimize power consumption from the power-hungry gain error calibration, the effect of VCO gain variation within the DPCM loop is investigated, leading to a calibration-free DPCM via a ΔΣ truncator. At the circuit level, a coarse-fine phase decoding scheme is proposed to increase signal-to-quantization-noise ratio (SQNR) while avoiding metastability issues. Fabricated in 65 nm, the AFE occupies an area of 0.056 mm². It achieves 94.2-dB SNDR and 95.1-dB DR in 500 Hz of bandwidth while dissipating 1.68-μW power from a single 0.8-V supply, yielding a Schreier FoM of 178.9 dB. Importantly, the techniques enable a virtually distortion-less spectrum with a noise-floor limited spurious-free dynamic range (SFDR) of 128 dB. In addition, the AFE is validated with ECG and EOG recording, demonstrating complete signal capturing in the presence of motion/stimulation artifacts.
8 October 2021 in IEEE Journal of Solid-State Circuits
This article presents a low-power (LP) area-efficient implantable neural recording system that supports high-density neural implant (HDNI) applications. The system uses a time-division multiple access method to record from 16-neural electrodes simultaneously. A least mean squares (LMSs) algorithm is used to cancel the slowly varying electrode offsets from all channels simultaneously by using a single-tap digital adaptive filter (AF). The presented technique is fabricated in 65-nm CMOS technology and achieves a per-channel area of 0.00248 mm 2 ; 68% of which is digital circuitry (and is thus scalable with technology). The overall system consumes 3.38 $\mu \text{W}$ per channel while achieving 2.6 $\mu \text{V}_{\mathrm {rms}}$ of input referred noise (IRN) in 10 kHz of bandwidth. The proposed system has a noise efficiency factor (NEF) of 1.83 and is fully integrated on-chip.
Next-generation wearable devices will enable clinical-grade, continuous ExG (ECG, EEG, EMG, etc.) biopotential monitoring, providing medical professionals with valuable longitudinal data outside of hospital settings. These devices must be ultra-low power $(\lt 10 \mu W)$ to enable long battery life while accurately digitizing sub-kHz, $\mu V-$level ExG signals in the presence of large motion and/or stimulation artifacts $(\gt 100$ mV) with high input-impedance $(Z_{in}\gt 10M\Omega)$ to avoid signal attenuation. Achieving such performance with conventional PGA + ADC architectures is challenging due to the conflicting low-power and $\gt 90$ dB dynamic-range (DR) requirements [1]. To address this, several direct digitization analog front-ends (AFEs) have been reported [2]–[4]. While these offer wide DR, they typically have low input-impedance $(\lt 5 M\Omega)$ and/or low power-efficiency (FoM $\le 172$ dB). This paper presents a scalable $2^{nd}-$ order voltage-controlled oscillator (VCO)-only $\Delta \sum$ ADC that achieves 92.3dB SNDR in a 1kHz bandwidth using a mismatch tolerant, multiphase gated-inverted ring-oscillator (GIRO) quantizer with dynamic power-scaling. The ADC uses an impedance-booster to maintain $\gt 50 M\Omega$ input-impedance over the entire bandwidth while consuming $4.25 \mu W$ during nominal operation and $5.8 \mu W$ in the presence of artifacts resulting in an FoM of 174.7dB.