VLSI architectures for Weighted Order Statistic (WOS) filters

Chaitali Chakrabarti, Lori E. Lucke

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

The class of median filters has been extended to include weighted order statistics (WOS) filters, to improve the flexibility of the filtering operation. The WOS filter weights each input within a sample window, and thus retains the original temporal order information. In this paper, we present efficient VLSI architectures for WOS filters which maintain a weighted rank for each sample in the sample window and update the weighted ranks for each window shift. We present novel (i) array architectures, (ii) stack filter architectures and (iii) sorting network architectures for non-recursive and recursive WOS filters which implement the above procedure. Our analysis shows that the bit-serial stack filter implementation is the one with smallest area while the bit-parallel stack filter is the one with the smallest input-output latency. The sorting network architecture (based on updating a sorted list) has the best area-time performance. Physical implementations verify our analysis.

Original languageEnglish (US)
Pages (from-to)1419-1433
Number of pages15
JournalSignal Processing
Volume80
Issue number8
DOIs
StatePublished - Aug 2000

Fingerprint

Statistics
Network architecture
Sorting
Median filters

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

VLSI architectures for Weighted Order Statistic (WOS) filters. / Chakrabarti, Chaitali; Lucke, Lori E.

In: Signal Processing, Vol. 80, No. 8, 08.2000, p. 1419-1433.

Research output: Contribution to journalArticle

@article{bf730c6f3b204ea5bd18754ebed98669,
title = "VLSI architectures for Weighted Order Statistic (WOS) filters",
abstract = "The class of median filters has been extended to include weighted order statistics (WOS) filters, to improve the flexibility of the filtering operation. The WOS filter weights each input within a sample window, and thus retains the original temporal order information. In this paper, we present efficient VLSI architectures for WOS filters which maintain a weighted rank for each sample in the sample window and update the weighted ranks for each window shift. We present novel (i) array architectures, (ii) stack filter architectures and (iii) sorting network architectures for non-recursive and recursive WOS filters which implement the above procedure. Our analysis shows that the bit-serial stack filter implementation is the one with smallest area while the bit-parallel stack filter is the one with the smallest input-output latency. The sorting network architecture (based on updating a sorted list) has the best area-time performance. Physical implementations verify our analysis.",
author = "Chaitali Chakrabarti and Lucke, {Lori E.}",
year = "2000",
month = "8",
doi = "10.1016/S0165-1684(00)00046-3",
language = "English (US)",
volume = "80",
pages = "1419--1433",
journal = "Signal Processing",
issn = "0165-1684",
publisher = "Elsevier",
number = "8",

}

TY - JOUR

T1 - VLSI architectures for Weighted Order Statistic (WOS) filters

AU - Chakrabarti, Chaitali

AU - Lucke, Lori E.

PY - 2000/8

Y1 - 2000/8

N2 - The class of median filters has been extended to include weighted order statistics (WOS) filters, to improve the flexibility of the filtering operation. The WOS filter weights each input within a sample window, and thus retains the original temporal order information. In this paper, we present efficient VLSI architectures for WOS filters which maintain a weighted rank for each sample in the sample window and update the weighted ranks for each window shift. We present novel (i) array architectures, (ii) stack filter architectures and (iii) sorting network architectures for non-recursive and recursive WOS filters which implement the above procedure. Our analysis shows that the bit-serial stack filter implementation is the one with smallest area while the bit-parallel stack filter is the one with the smallest input-output latency. The sorting network architecture (based on updating a sorted list) has the best area-time performance. Physical implementations verify our analysis.

AB - The class of median filters has been extended to include weighted order statistics (WOS) filters, to improve the flexibility of the filtering operation. The WOS filter weights each input within a sample window, and thus retains the original temporal order information. In this paper, we present efficient VLSI architectures for WOS filters which maintain a weighted rank for each sample in the sample window and update the weighted ranks for each window shift. We present novel (i) array architectures, (ii) stack filter architectures and (iii) sorting network architectures for non-recursive and recursive WOS filters which implement the above procedure. Our analysis shows that the bit-serial stack filter implementation is the one with smallest area while the bit-parallel stack filter is the one with the smallest input-output latency. The sorting network architecture (based on updating a sorted list) has the best area-time performance. Physical implementations verify our analysis.

UR - http://www.scopus.com/inward/record.url?scp=0034250797&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034250797&partnerID=8YFLogxK

U2 - 10.1016/S0165-1684(00)00046-3

DO - 10.1016/S0165-1684(00)00046-3

M3 - Article

AN - SCOPUS:0034250797

VL - 80

SP - 1419

EP - 1433

JO - Signal Processing

JF - Signal Processing

SN - 0165-1684

IS - 8

ER -