2-D Discrete Cosine Transform (DCT) on meshes with hierarchical control modes

Cheong Ghil Kim, Su Jin Lee, Shin Dug Kim

Research output: Contribution to journalConference article

2 Citations (Scopus)

Abstract

An effective matrix operation is critical to process 2-D DCT. This paper presents a hierarchically controlled SIMD array (HCSA) well suited to matrix computations, in which a conventional 2-D torus is enhanced with the hierarchical organization of control units and the global data buses running across the rows and columns. The distinguished features of the HCSA are the diagonally indexed concurrent broadcast and the efficient data exchanges among PEs through either row or column broadcast. Therefore, the HCSA can provide significant improvement on computation steps of DCT. For the performance evaluation, an algorithmic mapping method is used and the number of computation steps is analytically compared with semisystolic architecture.

Original languageEnglish
Pages (from-to)675-682
Number of pages8
JournalLecture Notes in Computer Science
Volume3522
Issue numberI
Publication statusPublished - 2005 Oct 3

Fingerprint

Hierarchical Control
Discrete Cosine Transform
Discrete cosine transforms
Mesh
Broadcast
Matrix Computation
Data Exchange
Electronic data interchange
Performance Evaluation
Concurrent
Torus
Unit

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

@article{eedeb7fe45154730a71d1615e63f9929,
title = "2-D Discrete Cosine Transform (DCT) on meshes with hierarchical control modes",
abstract = "An effective matrix operation is critical to process 2-D DCT. This paper presents a hierarchically controlled SIMD array (HCSA) well suited to matrix computations, in which a conventional 2-D torus is enhanced with the hierarchical organization of control units and the global data buses running across the rows and columns. The distinguished features of the HCSA are the diagonally indexed concurrent broadcast and the efficient data exchanges among PEs through either row or column broadcast. Therefore, the HCSA can provide significant improvement on computation steps of DCT. For the performance evaluation, an algorithmic mapping method is used and the number of computation steps is analytically compared with semisystolic architecture.",
author = "Kim, {Cheong Ghil} and Lee, {Su Jin} and Kim, {Shin Dug}",
year = "2005",
month = "10",
day = "3",
language = "English",
volume = "3522",
pages = "675--682",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",
number = "I",

}

2-D Discrete Cosine Transform (DCT) on meshes with hierarchical control modes. / Kim, Cheong Ghil; Lee, Su Jin; Kim, Shin Dug.

In: Lecture Notes in Computer Science, Vol. 3522, No. I, 03.10.2005, p. 675-682.

Research output: Contribution to journalConference article

TY - JOUR

T1 - 2-D Discrete Cosine Transform (DCT) on meshes with hierarchical control modes

AU - Kim, Cheong Ghil

AU - Lee, Su Jin

AU - Kim, Shin Dug

PY - 2005/10/3

Y1 - 2005/10/3

N2 - An effective matrix operation is critical to process 2-D DCT. This paper presents a hierarchically controlled SIMD array (HCSA) well suited to matrix computations, in which a conventional 2-D torus is enhanced with the hierarchical organization of control units and the global data buses running across the rows and columns. The distinguished features of the HCSA are the diagonally indexed concurrent broadcast and the efficient data exchanges among PEs through either row or column broadcast. Therefore, the HCSA can provide significant improvement on computation steps of DCT. For the performance evaluation, an algorithmic mapping method is used and the number of computation steps is analytically compared with semisystolic architecture.

AB - An effective matrix operation is critical to process 2-D DCT. This paper presents a hierarchically controlled SIMD array (HCSA) well suited to matrix computations, in which a conventional 2-D torus is enhanced with the hierarchical organization of control units and the global data buses running across the rows and columns. The distinguished features of the HCSA are the diagonally indexed concurrent broadcast and the efficient data exchanges among PEs through either row or column broadcast. Therefore, the HCSA can provide significant improvement on computation steps of DCT. For the performance evaluation, an algorithmic mapping method is used and the number of computation steps is analytically compared with semisystolic architecture.

UR - http://www.scopus.com/inward/record.url?scp=25144523517&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=25144523517&partnerID=8YFLogxK

M3 - Conference article

VL - 3522

SP - 675

EP - 682

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

IS - I

ER -