Realization of K-Nearest-Matches Search Capability in Fully-Parallel Associative Memories

Md. Anwarul ABEDIN; Yuki TANAKA; Ali AHMADI; Shogo SAKAKIBARA; Tetsushi KOIDE; Hans Jurgen MATTAUSCH

doi:10.1093/ietfec/e90-a.6.1240

Realization of K-Nearest-Matches Search Capability in Fully-Parallel Associative Memories

Md. Anwarul ABEDIN, Yuki TANAKA, Ali AHMADI, Shogo SAKAKIBARA, Tetsushi KOIDE, Hans Jurgen MATTAUSCH

Full Text Views

0

Share
Cite this

Summary :

The realization of k-nearest-matches search capability in fully-parallel mixed digital-analog associative memories by a sequential autonomous search mode is reported. The proposed concept and circuit implementation can be applied with all types of distance measures such as Hamming, Manhattan or Euclidean distance search, and the k value can be freely selected during operation. A test chip for concept verification has been designed in 0.35 µm CMOS technology with two-poly, three-metal layers, realizes k-nearest-matches Euclidean distance search and consumes 5.12 mm² of the chip area for 64 reference patterns each with 16 units of 5-bit.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E90-A No.6 pp.1240-1243

Publication Date: 2007/06/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1093/ietfec/e90-a.6.1240

Type of Manuscript: LETTER

Category: VLSI Design Technology and CAD

Cite this

Copy

Md. Anwarul ABEDIN, Yuki TANAKA, Ali AHMADI, Shogo SAKAKIBARA, Tetsushi KOIDE, Hans Jurgen MATTAUSCH, "Realization of K-Nearest-Matches Search Capability in Fully-Parallel Associative Memories" in IEICE TRANSACTIONS on Fundamentals, vol. E90-A, no. 6, pp. 1240-1243, June 2007, doi: 10.1093/ietfec/e90-a.6.1240.
Abstract: The realization of k-nearest-matches search capability in fully-parallel mixed digital-analog associative memories by a sequential autonomous search mode is reported. The proposed concept and circuit implementation can be applied with all types of distance measures such as Hamming, Manhattan or Euclidean distance search, and the k value can be freely selected during operation. A test chip for concept verification has been designed in 0.35 µm CMOS technology with two-poly, three-metal layers, realizes k-nearest-matches Euclidean distance search and consumes 5.12 mm² of the chip area for 64 reference patterns each with 16 units of 5-bit.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1093/ietfec/e90-a.6.1240/_p

Copy

@ARTICLE{e90-a_6_1240,
author={Md. Anwarul ABEDIN, Yuki TANAKA, Ali AHMADI, Shogo SAKAKIBARA, Tetsushi KOIDE, Hans Jurgen MATTAUSCH, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Realization of K-Nearest-Matches Search Capability in Fully-Parallel Associative Memories},
year={2007},
volume={E90-A},
number={6},
pages={1240-1243},
abstract={The realization of k-nearest-matches search capability in fully-parallel mixed digital-analog associative memories by a sequential autonomous search mode is reported. The proposed concept and circuit implementation can be applied with all types of distance measures such as Hamming, Manhattan or Euclidean distance search, and the k value can be freely selected during operation. A test chip for concept verification has been designed in 0.35 µm CMOS technology with two-poly, three-metal layers, realizes k-nearest-matches Euclidean distance search and consumes 5.12 mm² of the chip area for 64 reference patterns each with 16 units of 5-bit.},
keywords={},
doi={10.1093/ietfec/e90-a.6.1240},
ISSN={1745-1337},
month={June},}

Copy

TY - JOUR
TI - Realization of K-Nearest-Matches Search Capability in Fully-Parallel Associative Memories
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1240
EP - 1243
AU - Md. Anwarul ABEDIN
AU - Yuki TANAKA
AU - Ali AHMADI
AU - Shogo SAKAKIBARA
AU - Tetsushi KOIDE
AU - Hans Jurgen MATTAUSCH
PY - 2007
DO - 10.1093/ietfec/e90-a.6.1240
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E90-A
IS - 6
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - June 2007
AB - The realization of k-nearest-matches search capability in fully-parallel mixed digital-analog associative memories by a sequential autonomous search mode is reported. The proposed concept and circuit implementation can be applied with all types of distance measures such as Hamming, Manhattan or Euclidean distance search, and the k value can be freely selected during operation. A test chip for concept verification has been designed in 0.35 µm CMOS technology with two-poly, three-metal layers, realizes k-nearest-matches Euclidean distance search and consumes 5.12 mm² of the chip area for 64 reference patterns each with 16 units of 5-bit.
ER -