Moves like Jagger: Exploiting variations in instantaneous gait for spontaneous device pairing

Research output: Contribution to journalArticleScientificpeer-review

Standard

Moves like Jagger : Exploiting variations in instantaneous gait for spontaneous device pairing. / Schürmann, Dominik; Brüsch, Arne; Nguyen, Ngu; Sigg, Stephan; Wolf, Lars.

In: Pervasive and Mobile Computing, Vol. 47, 01.07.2018, p. 1-12.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

APA

Vancouver

Author

Bibtex - Download

@article{932af572ad1d485094876a67204bf531,
title = "Moves like Jagger: Exploiting variations in instantaneous gait for spontaneous device pairing",
abstract = "Seamless device pairing conditioned on the context of use fosters novel application domains and ease of use. Examples are automatic device pairings with objects interacted with, such as instrumented shopping baskets, electronic tourist guides (e.g. tablets), fitness trackers or other fitness equipment. We propose a cryptographically secure spontaneous authentication scheme, BANDANA, that exploits correlation in acceleration sequences from devices worn or carried together by the same person to extract always-fresh secure secrets. On two real world datasets with 15 and 482 subjects, BANDANA generated fingerprints achieved intra- (50{\%}) and inter-body (>75{\%}) similarity sufficient for secure key generation via fuzzy cryptography. Using BCH codes, best results are achieved with 48 bit fingerprints from 12 gait cycles generating 16 bit long keys. Statistical bias of the generated fingerprints has been evaluated as well as vulnerabilities towards relevant attack scenarios.",
keywords = "ad-hoc secure pairing, Authentication, Fuzzy cryptography, Gait",
author = "Dominik Sch{\"u}rmann and Arne Br{\"u}sch and Ngu Nguyen and Stephan Sigg and Lars Wolf",
year = "2018",
month = "7",
day = "1",
doi = "10.1016/j.pmcj.2018.03.006",
language = "English",
volume = "47",
pages = "1--12",
journal = "Pervasive and Mobile Computing",
issn = "1574-1192",
publisher = "Elsevier",

}

RIS - Download

TY - JOUR

T1 - Moves like Jagger

T2 - Exploiting variations in instantaneous gait for spontaneous device pairing

AU - Schürmann, Dominik

AU - Brüsch, Arne

AU - Nguyen, Ngu

AU - Sigg, Stephan

AU - Wolf, Lars

PY - 2018/7/1

Y1 - 2018/7/1

N2 - Seamless device pairing conditioned on the context of use fosters novel application domains and ease of use. Examples are automatic device pairings with objects interacted with, such as instrumented shopping baskets, electronic tourist guides (e.g. tablets), fitness trackers or other fitness equipment. We propose a cryptographically secure spontaneous authentication scheme, BANDANA, that exploits correlation in acceleration sequences from devices worn or carried together by the same person to extract always-fresh secure secrets. On two real world datasets with 15 and 482 subjects, BANDANA generated fingerprints achieved intra- (50%) and inter-body (>75%) similarity sufficient for secure key generation via fuzzy cryptography. Using BCH codes, best results are achieved with 48 bit fingerprints from 12 gait cycles generating 16 bit long keys. Statistical bias of the generated fingerprints has been evaluated as well as vulnerabilities towards relevant attack scenarios.

AB - Seamless device pairing conditioned on the context of use fosters novel application domains and ease of use. Examples are automatic device pairings with objects interacted with, such as instrumented shopping baskets, electronic tourist guides (e.g. tablets), fitness trackers or other fitness equipment. We propose a cryptographically secure spontaneous authentication scheme, BANDANA, that exploits correlation in acceleration sequences from devices worn or carried together by the same person to extract always-fresh secure secrets. On two real world datasets with 15 and 482 subjects, BANDANA generated fingerprints achieved intra- (50%) and inter-body (>75%) similarity sufficient for secure key generation via fuzzy cryptography. Using BCH codes, best results are achieved with 48 bit fingerprints from 12 gait cycles generating 16 bit long keys. Statistical bias of the generated fingerprints has been evaluated as well as vulnerabilities towards relevant attack scenarios.

KW - ad-hoc secure pairing

KW - Authentication

KW - Fuzzy cryptography

KW - Gait

UR - http://www.scopus.com/inward/record.url?scp=85045704163&partnerID=8YFLogxK

U2 - 10.1016/j.pmcj.2018.03.006

DO - 10.1016/j.pmcj.2018.03.006

M3 - Article

AN - SCOPUS:85045704163

VL - 47

SP - 1

EP - 12

JO - Pervasive and Mobile Computing

JF - Pervasive and Mobile Computing

SN - 1574-1192

ER -

ID: 19264733