Implementation and evaluation of an FPGA-based network data anonymizer

Yuichi Nakamura, Sota Sawaguchi, Hiroaki Nishi

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Privacy concern is a key element in recent Internet development, as more data transfers are acquired and analyzed while packets traverse the network. Data analysis can provide many benefits to internet service providers and users, such as recommendation services based on user activity on the Internet. While these services make use of the vast amounts of content and statistical data that can be acquired from network transactions, service providers face the challenge of ensuring that their services do not cause any privacy infringements. To cope with this problem, anonymization methods such as anonymity and l-diversity are commonly utilized to generalize datasets and eliminate private information. However, performing anonymization on a vast amount of data acquired from a network imposes high calculation costs. In this paper, we focus on proposing and implementing a field-programmable gate array (FPGA)-based data anonymizer to accelerate the anonymization of data that are used for internet services. A RAM-based anonymizer is proposed and tested using real-life internet traffic to evaluate how it performs under network applications.

Original languageEnglish
Pages (from-to)S134-S140
JournalIEEJ Transactions on Electrical and Electronic Engineering
Volume12
DOIs
Publication statusPublished - 2017 Jun 1
Externally publishedYes

Fingerprint

Field programmable gate arrays (FPGA)
Internet
Internet service providers
Random access storage
Data transfer
Costs

Keywords

  • anonymization
  • FPGA
  • k-anonymity
  • l-diversity

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Implementation and evaluation of an FPGA-based network data anonymizer. / Nakamura, Yuichi; Sawaguchi, Sota; Nishi, Hiroaki.

In: IEEJ Transactions on Electrical and Electronic Engineering, Vol. 12, 01.06.2017, p. S134-S140.

Research output: Contribution to journalArticle

@article{8841f779e5934d71b7e8606b8ae6bff3,
title = "Implementation and evaluation of an FPGA-based network data anonymizer",
abstract = "Privacy concern is a key element in recent Internet development, as more data transfers are acquired and analyzed while packets traverse the network. Data analysis can provide many benefits to internet service providers and users, such as recommendation services based on user activity on the Internet. While these services make use of the vast amounts of content and statistical data that can be acquired from network transactions, service providers face the challenge of ensuring that their services do not cause any privacy infringements. To cope with this problem, anonymization methods such as anonymity and l-diversity are commonly utilized to generalize datasets and eliminate private information. However, performing anonymization on a vast amount of data acquired from a network imposes high calculation costs. In this paper, we focus on proposing and implementing a field-programmable gate array (FPGA)-based data anonymizer to accelerate the anonymization of data that are used for internet services. A RAM-based anonymizer is proposed and tested using real-life internet traffic to evaluate how it performs under network applications.",
keywords = "anonymization, FPGA, k-anonymity, l-diversity",
author = "Yuichi Nakamura and Sota Sawaguchi and Hiroaki Nishi",
year = "2017",
month = "6",
day = "1",
doi = "10.1002/tee.22426",
language = "English",
volume = "12",
pages = "S134--S140",
journal = "IEEJ Transactions on Electrical and Electronic Engineering",
issn = "1931-4973",
publisher = "John Wiley and Sons Inc.",

}

TY - JOUR

T1 - Implementation and evaluation of an FPGA-based network data anonymizer

AU - Nakamura, Yuichi

AU - Sawaguchi, Sota

AU - Nishi, Hiroaki

PY - 2017/6/1

Y1 - 2017/6/1

N2 - Privacy concern is a key element in recent Internet development, as more data transfers are acquired and analyzed while packets traverse the network. Data analysis can provide many benefits to internet service providers and users, such as recommendation services based on user activity on the Internet. While these services make use of the vast amounts of content and statistical data that can be acquired from network transactions, service providers face the challenge of ensuring that their services do not cause any privacy infringements. To cope with this problem, anonymization methods such as anonymity and l-diversity are commonly utilized to generalize datasets and eliminate private information. However, performing anonymization on a vast amount of data acquired from a network imposes high calculation costs. In this paper, we focus on proposing and implementing a field-programmable gate array (FPGA)-based data anonymizer to accelerate the anonymization of data that are used for internet services. A RAM-based anonymizer is proposed and tested using real-life internet traffic to evaluate how it performs under network applications.

AB - Privacy concern is a key element in recent Internet development, as more data transfers are acquired and analyzed while packets traverse the network. Data analysis can provide many benefits to internet service providers and users, such as recommendation services based on user activity on the Internet. While these services make use of the vast amounts of content and statistical data that can be acquired from network transactions, service providers face the challenge of ensuring that their services do not cause any privacy infringements. To cope with this problem, anonymization methods such as anonymity and l-diversity are commonly utilized to generalize datasets and eliminate private information. However, performing anonymization on a vast amount of data acquired from a network imposes high calculation costs. In this paper, we focus on proposing and implementing a field-programmable gate array (FPGA)-based data anonymizer to accelerate the anonymization of data that are used for internet services. A RAM-based anonymizer is proposed and tested using real-life internet traffic to evaluate how it performs under network applications.

KW - anonymization

KW - FPGA

KW - k-anonymity

KW - l-diversity

UR - http://www.scopus.com/inward/record.url?scp=85020259884&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020259884&partnerID=8YFLogxK

U2 - 10.1002/tee.22426

DO - 10.1002/tee.22426

M3 - Article

VL - 12

SP - S134-S140

JO - IEEJ Transactions on Electrical and Electronic Engineering

JF - IEEJ Transactions on Electrical and Electronic Engineering

SN - 1931-4973

ER -