An MPEG-4 shape-adaptive inverse DCT with zero skipping and auto-aligned transpose memory

Hui Cheng Hsu*, Kun B. Lee, N. Y C Chang, Tian-Sheuan Chang

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

2 Scopus citations

Abstract

This paper presents an efficient VLSI architecture of shape-adaptive inverse discrete cosine transform (SA-IDCT) for the MPEG-4 video system. The proposed architecture contains a cost-effective 1-D variable-length IDCT engine and an auto-aligned transpose memory organization. The proposed design exploits the properties of SA-IDCT transform matrix to reduce area and improve throughput. Besides, the auto-aligned transpose memory organization can achieve transposing, shifting and aligning simultaneously with the capability of zero skipping. Compared to other designs, our architecture has higher throughout and lower power dissipation. When clocking at 62.5 MHz, the proposed architecture has a throughput of 401.41Mpixels/sec and power dissipation of 5.56 mW/sample @ 1.8V, 62.5MHz.

Original languageEnglish
Pages773-776
Number of pages4
StatePublished - 1 Dec 2004
Event2004 IEEE Asia-Pacific Conference on Circuits and Systems, APCCAS 2004: SoC Design for Ubiquitous Information Technology - Tainan, Taiwan
Duration: 6 Dec 20049 Dec 2004

Conference

Conference2004 IEEE Asia-Pacific Conference on Circuits and Systems, APCCAS 2004: SoC Design for Ubiquitous Information Technology
CountryTaiwan
CityTainan
Period6/12/049/12/04

Fingerprint Dive into the research topics of 'An MPEG-4 shape-adaptive inverse DCT with zero skipping and auto-aligned transpose memory'. Together they form a unique fingerprint.

Cite this