Design Of A Fully Predictive Ultra-low Power SAR ADC For Sparse Signals

In recent years,the gradual enhancement of personal health awareness has driven the rapid development of wearable smart devices and portable medical devices,and the demand for low power consumption of chips has increased.These devices help users to obtain their own physical health status through continuous detection of human bioelectrical signals.In order to obtain real-time physiological data,a signal acquisition system is required to continuously sample and quantify the user's bioelectrical signals.Therefore,it is particularly important to reduce power consumption and extend the life of the system.Bioelectric signal detection system usually consists of analog front-end circuit,ADC(analog-to-digital converter),digital signal processing circuit,RF transceiver circuit and other modules.Among them,the ADC,as the core module of the low-power analog detection circuit,has a great influence on the entire system.Therefore,designing an ADC capable of extracting the characteristics of a bioelectric signal,and the low-power ADC is very important for the entire system.Because SAR(successive approximation)ADC has the characteristics of low power consumption,simple structure,moderate speed accuracy,and easy integration,it is very suitable for this application scenario.In this paper,a 10-bit ultra-low-power SAR ADC suitable for sparse signals such as bioelectric signals is designed under the standard CMOS process of 0.13?m.Aiming at the sparse signals such as bioelectrical signals and sensor detection signals,the amplitude changes slowly for most of the time,the waveforms are clearly distinguished,and the signals change periodically with time.This paper proposes a quantization interval full prediction dynamic tracking algorithm.It can greatly reduce the average quantized power consumption of the low-frequency part of the signal,and at the same time,the quantized result of this algorithm can be used for subsequent bioelectrical signal characteristic parameter extraction and case detection.By combining the working principle of the SAR ADC and the full prediction algorithm,the DAC capacitor array corresponding to the algorithm and the overall circuit structure of the SAR ADC are designed.Through Matlab simulation modeling,the frequency,amplitude and other characteristics of bioelectrical signals are analyzed,and the feasibility of the algorithm and circuit structure is verified.The VCM-based segmented DAC structure is used to reduce the capacitor charging power consumption and the area of the DAC capacitor array.This article designs the circuit in the Virtuoso simulation environment,completes the layout design according to the circuit,and simulates the layout after parameter extraction.At the sampling rate of 10 kS/s,the post-simulation results based on Spectre/Hspice show that the spurious-free dynamic range(SFDR)is 66.7 dB,the signal-to-noise and distortion ratio(SNDR)is 58.4 dB,the effective number of bits(ENOB)is 9.4 bit,power consumption is 77.4 nW under 0.6 V power supply voltage.FoM value is 11.5 fJ/Conv.–s,and the total chip area is less than 1 mm~2.