CMOS图像传感器高一致性自适应斜坡电路设计方法

High consistency adaptive ramp circuit design for CMOS image sensors

  • 摘要:针对大面阵CMOS图像传感器(CIS)多斜坡发生器之间、多列之间由于后端布线寄生引起的非一致性问题,提出了一种基于分布式积分型的高一致性自适应斜坡电路设计方法。通过分析误差引入的根源,采用平均电压技术、自适应负反馈动态调节技术、数字相关双采样相结合的高精度补偿技术,完成了斜坡信号非一致性校准方案设计。实验结果表明,与现有的全局斜坡和分块式多斜坡相比,提出的斜坡产生电路DNL为+ 0.000636 LSB/− 0.0006 LSB,INL为+ 0.3292 LSB/− 0.7386 LSB,实现了斜坡信号的高精度;将各斜坡信号之间的不一致性降低为0.4% LSB,列固定模式噪声(CFPN)降低为 0.000037 %,而整个芯片仅增加了0.6%的面积和0.5%的功耗,该方法为超大面阵CMOS图像传感器斜坡信号的一致性提供了有效的解决方案。

    Abstract:This paper proposes a high consistency adaptive ramp circuit design method based on distributed integral type to address the inconsistency problem caused by parasitic backend wiring between multiple ramp generators and multiple columns in large-array CMOS image sensors (CIS). By analyzing the root causes of error introduction, a high-precision compensation technology combining average voltage technology, adaptive negative feedback dynamic adjustment technology, and digital correlation double sampling technology was adopted to complete the design of ramp signal inconsistency calibration scheme. The experimental results indicate that compared to existing global ramp and block-based multi-ramp approaches, the ramp generation circuit proposed in this paper achieves high accuracy of ramp signals with a DNL of + 0.000636 LSB/− 0.0006 LSB and an INL of + 0.3292 LSB/− 0.7386 LSB. The method reduces the inconsistency between the ramp signals to 0.4%LSB, decreases the column fixed pattern noise (CFPN) to 0.000037 %, and increases the overall chip area and power consumption by only 0.6% and 0.5%, respectively. this approach offers an effective solution for the consistency of ramp signals in large-scale array CMOS image sensors.

/

    返回文章
    返回
      Baidu
      map