Analysis of the matrix processing (M×P) architecture

Lakshmi Girija, Andreas Spanias

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we describe and evaluate a new DSP architecture called the matrix processing (M×P™). M×P exploits data parallelism using hardwired matrix operations and instructions. The M×P matrix computation capabilities are optimized for multidimensional transforms and filtering. A detailed analysis of the M×P performance, in terms of precision, execution time, and instruction memory requirements, is presented. We focus particularly on matrix and vector manipulations of the type embedded in video processing standards, e.g., filtering, DCT, and motion estimation. We present comparison tables of the performance of the MxP with other widely used DSPs such as the TI TMS320c55x™, and the TI TMS320c64x™.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2
DOIs
StatePublished - 2007
Event2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07 - Honolulu, HI, United States
Duration: Apr 15 2007Apr 20 2007

Other

Other2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07
CountryUnited States
CityHonolulu, HI
Period4/15/074/20/07

Fingerprint

matrices
Processing
education
discrete cosine transform
Motion estimation
manipulators
Data storage equipment
requirements

Keywords

  • DSP architectures
  • Video standards

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing
  • Acoustics and Ultrasonics

Cite this

Girija, L., & Spanias, A. (2007). Analysis of the matrix processing (M×P) architecture. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (Vol. 2). [4217424] https://doi.org/10.1109/ICASSP.2007.366251

Analysis of the matrix processing (M×P) architecture. / Girija, Lakshmi; Spanias, Andreas.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2 2007. 4217424.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Girija, L & Spanias, A 2007, Analysis of the matrix processing (M×P) architecture. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. vol. 2, 4217424, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07, Honolulu, HI, United States, 4/15/07. https://doi.org/10.1109/ICASSP.2007.366251
Girija L, Spanias A. Analysis of the matrix processing (M×P) architecture. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2. 2007. 4217424 https://doi.org/10.1109/ICASSP.2007.366251
Girija, Lakshmi ; Spanias, Andreas. / Analysis of the matrix processing (M×P) architecture. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2 2007.
@inproceedings{35ceeacaf16643b0ab09ff0f7fd43fce,
title = "Analysis of the matrix processing (M×P) architecture",
abstract = "In this paper, we describe and evaluate a new DSP architecture called the matrix processing (M×P™). M×P exploits data parallelism using hardwired matrix operations and instructions. The M×P matrix computation capabilities are optimized for multidimensional transforms and filtering. A detailed analysis of the M×P performance, in terms of precision, execution time, and instruction memory requirements, is presented. We focus particularly on matrix and vector manipulations of the type embedded in video processing standards, e.g., filtering, DCT, and motion estimation. We present comparison tables of the performance of the MxP with other widely used DSPs such as the TI TMS320c55x™, and the TI TMS320c64x™.",
keywords = "DSP architectures, Video standards",
author = "Lakshmi Girija and Andreas Spanias",
year = "2007",
doi = "10.1109/ICASSP.2007.366251",
language = "English (US)",
isbn = "1424407281",
volume = "2",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Analysis of the matrix processing (M×P) architecture

AU - Girija, Lakshmi

AU - Spanias, Andreas

PY - 2007

Y1 - 2007

N2 - In this paper, we describe and evaluate a new DSP architecture called the matrix processing (M×P™). M×P exploits data parallelism using hardwired matrix operations and instructions. The M×P matrix computation capabilities are optimized for multidimensional transforms and filtering. A detailed analysis of the M×P performance, in terms of precision, execution time, and instruction memory requirements, is presented. We focus particularly on matrix and vector manipulations of the type embedded in video processing standards, e.g., filtering, DCT, and motion estimation. We present comparison tables of the performance of the MxP with other widely used DSPs such as the TI TMS320c55x™, and the TI TMS320c64x™.

AB - In this paper, we describe and evaluate a new DSP architecture called the matrix processing (M×P™). M×P exploits data parallelism using hardwired matrix operations and instructions. The M×P matrix computation capabilities are optimized for multidimensional transforms and filtering. A detailed analysis of the M×P performance, in terms of precision, execution time, and instruction memory requirements, is presented. We focus particularly on matrix and vector manipulations of the type embedded in video processing standards, e.g., filtering, DCT, and motion estimation. We present comparison tables of the performance of the MxP with other widely used DSPs such as the TI TMS320c55x™, and the TI TMS320c64x™.

KW - DSP architectures

KW - Video standards

UR - http://www.scopus.com/inward/record.url?scp=34547519109&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547519109&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2007.366251

DO - 10.1109/ICASSP.2007.366251

M3 - Conference contribution

SN - 1424407281

SN - 9781424407286

VL - 2

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -