• DocumentCode
    184741
  • Title

    A 12.6 mW 8.3 Mevents/s contrast detection 128×128 imager with 75 dB intra-scene DR asynchronous random-access digital readout

  • Author

    Jongkil Park ; Sohmyung Ha ; Chul Kim ; Joshi, S. ; Yu, T. ; Wei Ma ; Cauwenberghs, G.

  • Author_Institution
    Univ. of California San Diego, La Jolla, CA, USA
  • fYear
    2014
  • fDate
    22-24 Oct. 2014
  • Firstpage
    564
  • Lastpage
    567
  • Abstract
    We present a temporal contrast detection event-driven asynchronous 128×128 pixel image sensor with integrated 10-bit ADC intensity readout in 0.18 μm 6M 1P CMOS. Current-mode log-compressed readout with correlated double sampling (CDS) offset compensation provides random access to instantaneous pixel intensity over 75 dB intra-scene dynamic range. Temporal correlated double sampling in “on” and “off” transient event detection reduces capacitor sizing requirements within the pixel. Global threshold adaptation using common-mode capacitive coupling across the pixel array compensates for global scene lighting variations. Power consumption at 8.3 Mevents/s is 12.6 mW from a 1.8 V supply, or 1.52 nJ per pixel event detection and readout. Die size is 5mm×5mm and pixel size is 33μm×33μm with 22.8% fill factor.
  • Keywords
    CMOS image sensors; biological techniques; power consumption; CDS offset compensation; CMOS; Mevent contrast detection imager; asynchronous pixel image sensor; common-mode capacitive coupling; correlated double sampling offset compensation; current-mode log-compressed readout; gain 75 dB; global scene lighting variations; integrated ADC intensity readout; intra-scene DR asynchronous random-access digital readout; pixel array compensation; pixel event detection; pixel event readout; power 12.6 mW; power consumption; temporal correlated double sampling; Arrays; Dynamic range; Event detection; Image sensors; Lighting; Retina; Sensors;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Biomedical Circuits and Systems Conference (BioCAS), 2014 IEEE
  • Conference_Location
    Lausanne
  • Type

    conf

  • DOI
    10.1109/BioCAS.2014.6981788
  • Filename
    6981788