节点文献

基于自适应相关多采样技术的CMOS图像传感器列级单斜ADC设计(英文)

A Single-Slope ADC with Adaptive Correlated Multiple Sampling Technique for CMOS Image Sensors

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 高静衡佳伟聂凯明徐江涛

【Author】 GAO Jing;HENG Jiawei;NIE Kaiming;XU Jiangtao;School of Microelectronics,Tianjin University;Tianjin Key Laboratory of Imaging and Sensing Microelectronic Technology;

【机构】 天津大学微电子学院天津市成像与感知微电子技术重点实验室

【摘要】 为了降低CMOS图像传感器的噪声水平,本文提出了一种基于自适应相关多采样技术的单斜ADC,可以根据光强自动选择合适的斜坡信号完成A/D转换。通过在低照度下使用小范围、高增益的斜坡信号进行多采样操作,可以充分降低读出噪声。多斜坡发生器由温度计码电容型DAC实现,此外还分析并校准了斜坡信号的失调电压和增益误差。提出的单斜ADC是在标准的110 nm 1P4M工艺下设计仿真的。仿真结果表明:基于自适应相关多采样技术的单斜ADC不会增加总A/D转换时间,同时在8倍的前置放大器增益下实现了最低59μVrms的读出噪声。此外当前置放大器为2倍增益时,读出噪声从强光下的211μVrms降低到弱光下的92μVrms,噪声的降低对应于7.21 dB信噪比的提高。另外,本文的品质因数为3.77 nVrms·s。

【Abstract】 This paper presents a single-slope ADC with adaptive correlated multiple sampling technique for lownoise imaging. Four ramps of different gains and durations are adaptively switched according to the light intensity.Therefore,the readout noise is reduced by using a small-range and high-gain ramp multiple times at low-light illumination. The multiple-ramp generator is based on the thermometer coded switch-capacitor array. Moreover,the offset voltage and gain error of the ramps are analyzed and calibrated. The single-slope ADC is implemented in a standard110 nm 1-poly 4-metal process. The simulation results show that the ADC with adaptive correlated multiple sampling technique achieves 59 μVrmsreadout noise at 8 × pre-amplifier gain without increasing the total A/D conversion time. Additionally,when the pre-amplifier gain is equal to 2,the readout noise is reduced from 211 μVrms to 92 μVrms,which corresponds to 7.21 dB signal-to-noise ratio improvement. Furthermore,the figure of merit of this paper is 3.77 nVrms·s.

【基金】 supported by the Tianjin Key Laboratory of Imaging and Sensing Microelectronic Technology
  • 【文献出处】 传感技术学报 ,Chinese Journal of Sensors and Actuators , 编辑部邮箱 ,2021年04期
  • 【分类号】TN792;TP212
  • 【下载频次】148
节点文献中: 

本文链接的文献网络图示:

本文的引文网络