A 2.7-M Pixels 64-mW CMOS Image Sensor with Multicolumn-Parallel Noise-Shaping SAR ADCs

Sun Il Hwang, Jae Hyun Chung, Hyeon June Kim, Il Hoon Jang, Min Jae Seo, Sang Hyun Cho, Heewon Kang, Minho Kwon, Seung Tak Ryu

Research output: Contribution to journalArticlepeer-review

20 Scopus citations

Abstract

This paper presents a CMOS image sensor (CIS) utilizing a noise-shaping successive-approximation register analog-to-digital converter (SAR ADC) incorporating the delta-readout scheme. While the noise-shaping SAR ADC with a proposed two-tap passive finite-impulse response (FIR) filter improves effective resolution, the delta-readout scheme reduces its power consumption. A prototype 1920× 1440 pixel CIS was fabricated in a 90-nm CIS process. A single-channel readout SAR ADC occupying an area of 22.4,μ m× 715,μ was implemented for reading out 16 columns of pixel array, consuming 437 μ W. Owing to the proposed noise-shaping SAR ADC with oversampling ratio of 16, this paper achieves a noise reduction of 14 dB compared with the noise of a conventional SAR ADC. The delta-readout reduces the power consumption of the SAR ADC by 10% due to the high hit rate of the full high definition image format. The measured differential nonlinearity of the ADC is +0.77/-0.54 LSB and the integral nonlinearity is +0.81/-0.5 LSB. The prototype CIS consumes a total power of 64 mW and achieves a dynamic range of 66.5 dB and a figure of merit of 127 μ VnJ at a data rate of 138 Mpixels/s.

Original languageEnglish
Pages (from-to)1119-1126
Number of pages8
JournalIEEE Transactions on Electron Devices
Volume65
Issue number3
DOIs
StatePublished - Mar 2018

Keywords

  • CMOS image sensor (CIS)
  • delta-readout scheme
  • noise-shaping successive-approximation register analog-to-digital converter (SAR ADC)
  • oversampling

Fingerprint

Dive into the research topics of 'A 2.7-M Pixels 64-mW CMOS Image Sensor with Multicolumn-Parallel Noise-Shaping SAR ADCs'. Together they form a unique fingerprint.

Cite this