Defending against Deep-Learning-Based Flow Correlation Attacks with Adversarial Examples
Tor is vulnerable to flow correlation attacks, adversaries who can observe the traffic metadata (e.g., packet timing, size, etc.) between client to entry relay and exit relay to the server will deanonymize users by calculating the degree of association. A recent study has shown that deep-learning-ba...
Gespeichert in:
Veröffentlicht in: | Security and communication networks 2022-03, Vol.2022, p.1-11 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 11 |
---|---|
container_issue | |
container_start_page | 1 |
container_title | Security and communication networks |
container_volume | 2022 |
creator | Zhang, Ziwei Ye, Dengpan |
description | Tor is vulnerable to flow correlation attacks, adversaries who can observe the traffic metadata (e.g., packet timing, size, etc.) between client to entry relay and exit relay to the server will deanonymize users by calculating the degree of association. A recent study has shown that deep-learning-based approach called DeepCorr provides a high flow correlation accuracy of over 96%. The escalating threat of this attack requires timely and effective countermeasures. In this paper, we propose a novel defense mechanism that injects dummy packets into flow traces by precomputing adversarial examples, successfully breaks the flow pattern that CNNs model has learned, and achieves a high protection success rate of over 97%. Moreover, our defense only requires 20% bandwidth overhead, which outperforms the state-of-the-art defense. We further consider implementing our defense in the real world. We find that, unlike traditional scenarios, the traffic flows are “fixed” only when they are coming, which means we must know the next packet’s feature. In addition, the websites are not immutable, and the characteristics of the transmitted packets will change irregularly and lead to the inefficiency of adversarial samples. To solve these problems, we design a system to adapt our defense in the real world and further reduce bandwidth overhead. |
doi_str_mv | 10.1155/2022/2962318 |
format | Article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2646640489</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2646640489</sourcerecordid><originalsourceid>FETCH-LOGICAL-c294t-de385ed79e04e85086d8bfebeb4ab71ae9b326a391085d802e79fa9195fef2e33</originalsourceid><addsrcrecordid>eNp9kF1LwzAUhoMoOKd3_oCAl1qXpGmbXM59qDDwRsG7crqcbpldWpPO6r-3Y8NLr87L4eF94SHkmrN7zpNkJJgQI6FTEXN1QgZcxzpiXIjTv8zlObkIYcNYymUmB-R9iiU6Y92KwgqsCy2dIjbRAsG7_hs9QEBD51Xd0UntPVbQ2trRcdvC8iPQzrZrOjZf6AN4CxWdfcO2qTBckrMSqoBXxzskb_PZ6-QpWrw8Pk_Gi2gptGwjg7FK0GQamUSVMJUaVZRYYCGhyDigLmKRQqw5U4lRTGCmS9BcJyWWAuN4SG4OvY2vP3cY2nxT77zrJ3ORyjSVTCrdU3cHaunrEDyWeePtFvxPzlm-d5fv3eVHdz1-e8DX1hno7P_0L_LQblk</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2646640489</pqid></control><display><type>article</type><title>Defending against Deep-Learning-Based Flow Correlation Attacks with Adversarial Examples</title><source>Wiley-Blackwell Open Access Titles(OpenAccess)</source><source>EZB-FREE-00999 freely available EZB journals</source><source>Alma/SFX Local Collection</source><creator>Zhang, Ziwei ; Ye, Dengpan</creator><contributor>Alazab, Mamoun ; Mamoun Alazab</contributor><creatorcontrib>Zhang, Ziwei ; Ye, Dengpan ; Alazab, Mamoun ; Mamoun Alazab</creatorcontrib><description>Tor is vulnerable to flow correlation attacks, adversaries who can observe the traffic metadata (e.g., packet timing, size, etc.) between client to entry relay and exit relay to the server will deanonymize users by calculating the degree of association. A recent study has shown that deep-learning-based approach called DeepCorr provides a high flow correlation accuracy of over 96%. The escalating threat of this attack requires timely and effective countermeasures. In this paper, we propose a novel defense mechanism that injects dummy packets into flow traces by precomputing adversarial examples, successfully breaks the flow pattern that CNNs model has learned, and achieves a high protection success rate of over 97%. Moreover, our defense only requires 20% bandwidth overhead, which outperforms the state-of-the-art defense. We further consider implementing our defense in the real world. We find that, unlike traditional scenarios, the traffic flows are “fixed” only when they are coming, which means we must know the next packet’s feature. In addition, the websites are not immutable, and the characteristics of the transmitted packets will change irregularly and lead to the inefficiency of adversarial samples. To solve these problems, we design a system to adapt our defense in the real world and further reduce bandwidth overhead.</description><identifier>ISSN: 1939-0114</identifier><identifier>EISSN: 1939-0122</identifier><identifier>DOI: 10.1155/2022/2962318</identifier><language>eng</language><publisher>London: Hindawi</publisher><subject>Accuracy ; Bandwidths ; Classification ; Communications traffic ; Correlation ; Deep learning ; Flow distribution ; Noise ; Packet transmission ; Relay ; Success ; Traffic flow ; Web sites ; Websites</subject><ispartof>Security and communication networks, 2022-03, Vol.2022, p.1-11</ispartof><rights>Copyright © 2022 Ziwei Zhang and Dengpan Ye.</rights><rights>Copyright © 2022 Ziwei Zhang and Dengpan Ye. This is an open access article distributed under the Creative Commons Attribution License (the “License”), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. https://creativecommons.org/licenses/by/4.0</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c294t-de385ed79e04e85086d8bfebeb4ab71ae9b326a391085d802e79fa9195fef2e33</cites><orcidid>0000-0002-2320-6728 ; 0000-0003-2510-9523</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><contributor>Alazab, Mamoun</contributor><contributor>Mamoun Alazab</contributor><creatorcontrib>Zhang, Ziwei</creatorcontrib><creatorcontrib>Ye, Dengpan</creatorcontrib><title>Defending against Deep-Learning-Based Flow Correlation Attacks with Adversarial Examples</title><title>Security and communication networks</title><description>Tor is vulnerable to flow correlation attacks, adversaries who can observe the traffic metadata (e.g., packet timing, size, etc.) between client to entry relay and exit relay to the server will deanonymize users by calculating the degree of association. A recent study has shown that deep-learning-based approach called DeepCorr provides a high flow correlation accuracy of over 96%. The escalating threat of this attack requires timely and effective countermeasures. In this paper, we propose a novel defense mechanism that injects dummy packets into flow traces by precomputing adversarial examples, successfully breaks the flow pattern that CNNs model has learned, and achieves a high protection success rate of over 97%. Moreover, our defense only requires 20% bandwidth overhead, which outperforms the state-of-the-art defense. We further consider implementing our defense in the real world. We find that, unlike traditional scenarios, the traffic flows are “fixed” only when they are coming, which means we must know the next packet’s feature. In addition, the websites are not immutable, and the characteristics of the transmitted packets will change irregularly and lead to the inefficiency of adversarial samples. To solve these problems, we design a system to adapt our defense in the real world and further reduce bandwidth overhead.</description><subject>Accuracy</subject><subject>Bandwidths</subject><subject>Classification</subject><subject>Communications traffic</subject><subject>Correlation</subject><subject>Deep learning</subject><subject>Flow distribution</subject><subject>Noise</subject><subject>Packet transmission</subject><subject>Relay</subject><subject>Success</subject><subject>Traffic flow</subject><subject>Web sites</subject><subject>Websites</subject><issn>1939-0114</issn><issn>1939-0122</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>RHX</sourceid><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GNUQQ</sourceid><recordid>eNp9kF1LwzAUhoMoOKd3_oCAl1qXpGmbXM59qDDwRsG7crqcbpldWpPO6r-3Y8NLr87L4eF94SHkmrN7zpNkJJgQI6FTEXN1QgZcxzpiXIjTv8zlObkIYcNYymUmB-R9iiU6Y92KwgqsCy2dIjbRAsG7_hs9QEBD51Xd0UntPVbQ2trRcdvC8iPQzrZrOjZf6AN4CxWdfcO2qTBckrMSqoBXxzskb_PZ6-QpWrw8Pk_Gi2gptGwjg7FK0GQamUSVMJUaVZRYYCGhyDigLmKRQqw5U4lRTGCmS9BcJyWWAuN4SG4OvY2vP3cY2nxT77zrJ3ORyjSVTCrdU3cHaunrEDyWeePtFvxPzlm-d5fv3eVHdz1-e8DX1hno7P_0L_LQblk</recordid><startdate>20220327</startdate><enddate>20220327</enddate><creator>Zhang, Ziwei</creator><creator>Ye, Dengpan</creator><general>Hindawi</general><general>Hindawi Limited</general><scope>RHU</scope><scope>RHW</scope><scope>RHX</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>8FD</scope><scope>8FE</scope><scope>8FG</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>GNUQQ</scope><scope>HCIFZ</scope><scope>JQ2</scope><scope>K7-</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>P5Z</scope><scope>P62</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><orcidid>https://orcid.org/0000-0002-2320-6728</orcidid><orcidid>https://orcid.org/0000-0003-2510-9523</orcidid></search><sort><creationdate>20220327</creationdate><title>Defending against Deep-Learning-Based Flow Correlation Attacks with Adversarial Examples</title><author>Zhang, Ziwei ; Ye, Dengpan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c294t-de385ed79e04e85086d8bfebeb4ab71ae9b326a391085d802e79fa9195fef2e33</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Accuracy</topic><topic>Bandwidths</topic><topic>Classification</topic><topic>Communications traffic</topic><topic>Correlation</topic><topic>Deep learning</topic><topic>Flow distribution</topic><topic>Noise</topic><topic>Packet transmission</topic><topic>Relay</topic><topic>Success</topic><topic>Traffic flow</topic><topic>Web sites</topic><topic>Websites</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Zhang, Ziwei</creatorcontrib><creatorcontrib>Ye, Dengpan</creatorcontrib><collection>Hindawi Publishing Complete</collection><collection>Hindawi Publishing Subscription Journals</collection><collection>Hindawi Publishing Open Access</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>Advanced Technologies & Aerospace Collection</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>ProQuest Central Student</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Computer Science Collection</collection><collection>Computer science database</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>ProQuest advanced technologies & aerospace journals</collection><collection>ProQuest Advanced Technologies & Aerospace Collection</collection><collection>ProQuest - Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><jtitle>Security and communication networks</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Zhang, Ziwei</au><au>Ye, Dengpan</au><au>Alazab, Mamoun</au><au>Mamoun Alazab</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Defending against Deep-Learning-Based Flow Correlation Attacks with Adversarial Examples</atitle><jtitle>Security and communication networks</jtitle><date>2022-03-27</date><risdate>2022</risdate><volume>2022</volume><spage>1</spage><epage>11</epage><pages>1-11</pages><issn>1939-0114</issn><eissn>1939-0122</eissn><abstract>Tor is vulnerable to flow correlation attacks, adversaries who can observe the traffic metadata (e.g., packet timing, size, etc.) between client to entry relay and exit relay to the server will deanonymize users by calculating the degree of association. A recent study has shown that deep-learning-based approach called DeepCorr provides a high flow correlation accuracy of over 96%. The escalating threat of this attack requires timely and effective countermeasures. In this paper, we propose a novel defense mechanism that injects dummy packets into flow traces by precomputing adversarial examples, successfully breaks the flow pattern that CNNs model has learned, and achieves a high protection success rate of over 97%. Moreover, our defense only requires 20% bandwidth overhead, which outperforms the state-of-the-art defense. We further consider implementing our defense in the real world. We find that, unlike traditional scenarios, the traffic flows are “fixed” only when they are coming, which means we must know the next packet’s feature. In addition, the websites are not immutable, and the characteristics of the transmitted packets will change irregularly and lead to the inefficiency of adversarial samples. To solve these problems, we design a system to adapt our defense in the real world and further reduce bandwidth overhead.</abstract><cop>London</cop><pub>Hindawi</pub><doi>10.1155/2022/2962318</doi><tpages>11</tpages><orcidid>https://orcid.org/0000-0002-2320-6728</orcidid><orcidid>https://orcid.org/0000-0003-2510-9523</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1939-0114 |
ispartof | Security and communication networks, 2022-03, Vol.2022, p.1-11 |
issn | 1939-0114 1939-0122 |
language | eng |
recordid | cdi_proquest_journals_2646640489 |
source | Wiley-Blackwell Open Access Titles(OpenAccess); EZB-FREE-00999 freely available EZB journals; Alma/SFX Local Collection |
subjects | Accuracy Bandwidths Classification Communications traffic Correlation Deep learning Flow distribution Noise Packet transmission Relay Success Traffic flow Web sites Websites |
title | Defending against Deep-Learning-Based Flow Correlation Attacks with Adversarial Examples |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T07%3A23%3A27IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Defending%20against%20Deep-Learning-Based%20Flow%20Correlation%20Attacks%20with%20Adversarial%20Examples&rft.jtitle=Security%20and%20communication%20networks&rft.au=Zhang,%20Ziwei&rft.date=2022-03-27&rft.volume=2022&rft.spage=1&rft.epage=11&rft.pages=1-11&rft.issn=1939-0114&rft.eissn=1939-0122&rft_id=info:doi/10.1155/2022/2962318&rft_dat=%3Cproquest_cross%3E2646640489%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2646640489&rft_id=info:pmid/&rfr_iscdi=true |