A capacitively-coupled chopper instrument amplifier (CCIA) for bio-potential monitoring analog front-end (AFE) with auto-calibrated dual positive feedback loops (DPFLs) is presented in this article. The proposed AFE with the DPFL does not only prevent total input impedance degradation by unexpected external and internal parasitic capacitances but also achieves excellent noise efficiency simultaneously. The DPFL consists of a conventional internal positive feedback loop with an attenuator for enhancing calibration resolution and an external positive feedback loop outside of the input chopper to compensate for a large external parasitic capacitance ( $C_{P-EXT}$ ), boosting total input impedance significantly. To precisely determine the feedback factor of the DPFL with respect to parasitic capacitors, an on-chip foreground calibration is also implemented in this work. Fabricated in a 110-nm CMOS process, the prototype AFE consumes 3.83 $\mu \text{W}$ per channel at 1 and 1.5-V supplies and has the input-referred noise of 0.36 $\mu \text{V}_{rms}$ from 0.5 to 300 Hz with the noise efficiency factor of 1.54. The proposed AFE achieves high-input impedances of 15 $\text{G}\Omega $ at 10 Hz and 2 $\text{G}\Omega $ at 50 Hz, even with an additional $C_{P-EXT}$ of 82 pF.
21 June 2022 in IEEE Journal of Solid-State Circuits
This article presents a bipolar junction transistor (BJT)-based CMOS temperature-to-frequency converter (TFC). A relaxation oscillator (ROSC) consisting of an integrating capacitor, a current source, and a comparator converts the ratio of the complementary-to-absolute-temperature (CTAT) voltage to the proportional-to-absolute-temperature (PTAT) voltage created by the BJT core to frequency. To improve the temperature accuracy, we incorporate a capacitor flipping technique to remove an offset of the comparator and a settling time error when resetting the integrating capacitor. Moreover, bootstrapped NMOS switches with the triple well structure are implemented to suppress leakage current and prevent from forming parasitic BJTs when the negative CTAT is applied to them by the proposed flipping technique. The prototype fabricated in a 0.11- $\mu \text{m}$ CMOS process consumes 3.1- $\mu \text{W}$ power and occupies a 0.025-mm 2 active area. Measurements of 18 samples with the digital trimming show an inaccuracy of $\pm 1~^{\circ }\text{C}$ ( $3\sigma$ ) from $- 40\,\,^{\circ }\text{C}$ to $140~^{\circ }\text{C}$ . The area- and energy-efficient TFC is applicable to monitor the temperature inside integrated chips.
3 October 2022 in IEEE Transactions on Biomedical Circuits and Systems
This article presents a CMOS microelectrode array (MEA) system with a reconfigurable sub-array multiplexing architecture using the time-division multiplexing (TDM) technique. The system consists of 24,320 TiN electrodes with 17.7 µm-pitch pixels and 380 column-parallel readout channels including a low-noise amplifier, a programmable gain amplifier, and a 10-b successive approximation register analog to digital converter. Readout channels are placed outside the pixel for high spatial resolution, and a flexible structure to acquire neural signals from electrodes selected by configuring in-pixel memory is realized. In this structure, a single channel can handle 8 to 32 electrodes, guaranteeing a temporal resolution from 5kS/s to 20kS/s for each electrode. A 128 × 190 MEA system was fabricated in a 110-nm CMOS process, and each readout channel consumes 81 µW at 1.5-V supply voltage featuring input-referred noise of 1.48 µV rms without multiplexing and 5.4 µV rms with multiplexing at the action-potential band (300 Hz – 10 kHz).
Monitoring the electrical neural signals is an important method for understanding the neuronal mechanism. In particular, in order to perform a cell-type-specific study, it is necessary to observe the concentration of calcium ions using fluorescent indicators in addition to measuring the electrical neural signal. This paper presents a multimodal multichannel neural activity readout integrated circuit that can perform not only electrical neural recording but also fluorescence recording of neural activity for the cell-type-specific study of heterogeneous neuronal cell populations. For monitoring the calcium ions, the photodiode generates the current according to the fluorescence expressed by the reaction between the genetically encoded calcium indicators and calcium ions. The time-based fluorescence recording circuit then records the photodiode current. The electrical neural signal captured by the microelectrode is recorded through the low-noise amplifier, variable gain amplifier, and analog-to-digital converter. The proposed integrated circuit is fabricated in a 1-poly 6-metal (1P6M) 0.18- $\mu \text{m}$ CMOS process. The fluorescence recording circuit achieves a recording range of 81 dB (75 pA to 860 nA) and consumes a power of 724 nW/channel. The electrical recording circuit achieves an input-referred noise of $2.7~\mu \text{V}_{\mathrm {rms}}$ over the bandwidth of 10 kHz, while consuming the power of $4.9~\mu \text{W}$ /channel. The functionality of the proposed circuits is verified through the in vivo and in vitro experiments. Compared to the conventional neuroscience tools, which consist of bulky off-chip components, this neural interface is implemented in a compact size to perform multimodal neural recording while consuming low power.
In this letter, a 2nd-order 2 -Σ modulator consisting of a voltage-controlled-oscillator-based quantizer (VCOQ) and a current digital-to-analog converter (DAC) with a pulse width modulator (PWM) is presented for the precise acquisition of a wide range photocurrent in an area-and energy-efficient form factor. The proposed 2-modulation realized by the 2nd-order infinite impulse response (IIR) filter on the feedback significantly attenuates the magnitude of input signals, enhancing the DR and linearity. Moreover, an additional differentiator followed by the VCOQ features the negative feedback loop in the 2nd-order Σ modulator, improving noise shaping with no additional current DAC noise. In addition, a 1-bit PWM current DAC substituting the multi-bit current DAC is devised to mitigate the noise from the current DAC, realizing the high resolution of 1 pA with 500 Hz bandwidth. The prototype chip fabricated in a 110-nm CMOS occupies 0.0308 mm2 and achieves Walden FoM of 4.15 pJ/conv.
4 May 2021 in IEEE Transactions on Circuits and Systems II: Express Briefs
We present a 32.768 kHz crystal oscillator with ultra-low power operation and a fast startup featuring for IoT applications. A self-reference power gating scheme detects the amplitude of a clock, enabling a negative resistance booster to awake XO and keep the oscillation with a duty-cycling technique for minimizing power consumption. In addition, a relaxation oscillator injects in-band noise initially, further reducing the startup time. The prototype XO for proof-of-concept was fabricated in 0.11- $\mu \text{m}$ CMOS and fully characterized. It consumes 5.02 nW at 0.5-V supply and achieves the startup time of 1.26 s simultaneously.