A 0.5V signal-specific continuous-time level-crossing ADC with charge sharing
A 0.5V signal-specific continuous-time level-crossing ADC with charge sharing
复制标题
具有电荷共享功能的 0.5V 信号特定连续时间电平交叉 ADC
DOI：
--
复制
发表时间：
2011
期刊：
影响因子：
--
通讯作者：
W. Serdijn
中科院分区：
文献类型：
--
作者：
Yongjia Li;Duan Zhao;Marijn van Dongen;W. Serdijn
针对生物医学信号传感应用，提出了一种新型的连续时间跨电平模数转换器(LC-ADC)。传统的数模转换器(DAC)被电荷共享模块取代，导致更低的功耗、更低的设计复杂性和更大的灵活性，适用于各种分辨率的应用。该ADC采用90 nm CMOS工艺设计，在0.5V和0.7V双电源供电下，最大信噪比为51.4分贝，最小功耗为538nW。模数转换器是可穿戴和植入式生物医学数据采集系统不可或缺的组成部分。传统的ADC使用均匀采样，该采样基于对信号幅度的周期性采样。在这些ADC结构中，逐次逼近寄存器(SAR)ADC是低功耗应用中最常用的结构，因为其中唯一的模拟电路是比较器，而其余电路是数字电路，因此很容易随技术缩小。基本上，SAR-ADC定期对输入进行采样，并使用二进制搜索算法从最高有效位(MSB)到最低有效位(LSB)依次逼近采样的输入信号。然而，对于生物医学信号记录，SAR-ADC并不是那么节能，因为大多数信号分量具有小的信号幅度，而即使在输入信号缓慢变化的情况下，SAR-ADC也针对每个输入样本重复从MSB到LSB的相同搜索过程。
This paper presents a novel continuous-time level- crossing analog-to-digital converter (LC-ADC) targeted at biomedical signal sensing applications. The conventional digital- to-analog converter (DAC) is replaced by a charge sharing block, leading to lower power consumption, less design complexity and more flexibility for various resolution applications. Designed to be implemented in 90 nm CMOS technology, the proposed ADC achieves a maximum SNDR of 51.4dB and consumes a minimum power of 538nW from a dual supply of 0.5V and 0.7V. I. INTRODUCTION Analog-to-digital converters are indispensable building blocks of wearable and implantable biomedical data acquisition systems. Conventional ADCs utilize uniform sampling, which is based on periodically sampling the magnitude of the signal. Among those ADC structures, the successive approximation register (SAR) ADC is the most popular one in low power applications, because the only analog circuitry within it is a comparator, while the rest of the circuits are digital and thus easily scale down with technology. Basically, SAR-ADCs periodically sample the input and use binary search algorithms to successively approximate the sampled input signal from the most-significant bit (MSB) to the least-significant bit (LSB). However, SAR-ADCs are not that power-efficient for biomedical signal recording, as most of the signal components have small signal magnitudes, while the SAR-ADC repeats the same searching procedure from MSB to LSB for each input sample even when the input signal varies slowly.