Skip to main navigation Skip to search Skip to main content

Design of high speed arithmetic encoder

  • Xi'an Jiaotong University

Research output: Contribution to conferencePaperpeer-review

7 Scopus citations

Abstract

Focusing on the problem of path waiting or circular that existed in updating of context table and the Renorme and Byteout procedure in the realization of the conventional arithmetic encoder in JPEG2000, 3-step pipeline architecture is used on FPGA to get high speed encoding. A method of updating CX table is proposed; and a circuit with short delay is also implemented to detect the left zeros of A-register. Multiplexers are adopted to accelerate the random left shift operation, and parallel processing based on data dependency is used to optimize RTL code to shorten the main critical path, the updating logic of context table is last fully discussed. Experimental result shows the encoder can work up to 107MHz on Altera's EP1S25B672C7 and the critical path is 4.6 ns synthesized in synopsys DC by TSMC.25um library.

Original languageEnglish
Pages1617-1620
Number of pages4
StatePublished - 2004
Event2004 7th International Conference on Solid-State and Integrated Circuits Technology Proceedings, ICSICT 2004 - Beijing, China
Duration: 18 Oct 200421 Oct 2004

Conference

Conference2004 7th International Conference on Solid-State and Integrated Circuits Technology Proceedings, ICSICT 2004
Country/TerritoryChina
CityBeijing
Period18/10/0421/10/04

Keywords

  • Arithmetic encoder
  • Critical path
  • JPEG2000
  • Pipeline

Fingerprint

Dive into the research topics of 'Design of high speed arithmetic encoder'. Together they form a unique fingerprint.

Cite this