Low-Power Parallel Video Compression Architecture for a Single-Chip Digital CMOS Camera

  • Authors:
  • Jeff Y. F. Hsieh;Teresa H. Y. Meng

  • Affiliations:
  • Center for Integrated Systems, Department of Electrical Engineering, Stanford University, Stanford CA 94305;Center for Integrated Systems, Department of Electrical Engineering, Stanford University, Stanford CA 94305

  • Venue:
  • Journal of VLSI Signal Processing Systems - Special issue on system level design
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

A low-power, large-scale parallel video compression architecture for asingle-chip digital CMOS camera is discussed in this paper. Thisarchitecture is designed for highly computationally intensive image andvideo processing tasks necessary to support video compression. Two designsof this architecture, an MPEG2 encoder and a DV encoder, are presented. Atan image resolution of 640 × 480 pixels (MPEG2) and 720 × 576 (DV)and a frame rate of 25 to 30 frames per second, a computational throughputof up to 1.8 billion operations per second (BOPS) is required. This issupported in the proposed architecture using a 40 MHz clock and an array of40 to 45 parallel processors implemented in a 0.2 μm CMOS technology andwith a 1.5 V supply voltage. Power consumption is significantly reducedthrough the single-chip integration of the CMOS photo sensors, the embeddedDRAM technology, and the proposed pipelined parallel processors. Theparallel processors consume approximately 45 mW of power resulting a powerefficiency of 40 BOPS/W.