PrivacyByDesign: Private Use of Biometrics EBF Research Seminar October 2

1
Privacy-By-Design Private Use of
BiometricsEBF Research SeminarOctober 2 3,
2007, BrusselsMichiel van der Veen
2
Agenda

• (Biometric) Identity Theft
• Privacy-by-Design
• privID for Private Speaker Identification
• Conclusion

3
Identity TheftA problem that affects everyone
and becomes more serious in the networked
environment

• Homeland security
• Fake IDs
• Terrorist threat
• Healthcare
• Insurance fraud
• Medical casualties
• Financial
• Credit card fraud, bank account transfer fraud
• 15 M victims (US only)
• Average loss per individual (US only)

2005
1,408 3,257
2006
source Gartner 2007
4
The ChallengeDoes biometric open up
possibilities for the ultimate identity theft ?

• Current biometric systems depend on perimeter
  based security solutions that are not guaranteed
  in a broad on-line deployment
• Biometric identity theft
• Unchangeable biometric over a persons
• lifetime (once compromised, compromised
• forever)

5
Agenda

• (Biometric) Identity Theft
• Privacy-by-Design
• privID for Private Speaker Identification
• Conclusion

6
Privacy-by-DesignCombating biometric identity
theft

• Fear for Identity Theft in Biometric Systems
  leading to a preference for local storage (11)
  versus centralized storage (1N), however
• The world is on-line and networked
• Limitations on application functionality and
  efficient data management
• Privacy-by-Design Guidelines
• Data minimization only store the necessary
  information
• Limited use principles
• Build in privacy up front, into the design
  specifications

A technology should reveal no more information
than is necessaryit should be built to be the
least revealing system possible. Lawrence
Lessig, 1999
7
Pseudo Biometric IdentitiesIntrinsically secured
biometric ID information
PrivID Template Protection

• Pseudo Biometric Identities
• Intrinsic security
• renewable and revocable

• Multi-biometric input
• Face, Finger, Iris, Voice

8
Privacy-by-Design in BiometricsKeeping the noise
down to generate unique and independent
identifiers
PrivID Template Protection
A-D Conversion
Error Correction
Hash/ Encrypt
Diversify
(i) Reduced Template Size
(ii) Exact Matching
(iii) Multiple Templates
(iv) Secured Templates
9
3-Stage Privacy Platform
10
Unlimited Applications and Uses of Template
Protection

• Biometric ticketing for events
• Biometric boarding cards for air travel
• Identification, credit and loyalty card systems
• Anonymous (untraceable) labeling of sensitive
  records (medical, financial)
• Consumer biometric payment systems
• Access control to personal computing devices
• Personal encryption products
• Local or remote authentication to access files
  held by government and other various
  organizations

Source Ann Cavoukian, Sept 17, 2007
11
Agenda

• (Biometric) Identity Theft
• Privacy-by-Design
• privID for Private Speaker Identification
• Conclusion

12
Private Speaker IdentificationWorlds first
pilot for biometric privacy-by-design solution

• Bell Canada is deploying voluntary voice identity
  verification service for its customers using
  PerSays world leading Voice Biometric technology
  - Sept 2007 325,000 voluntary enrollments
• Successful integration of privID with PerSays
  voice technology offering superior matching
  performance while including privacy
• Opportunity for worlds first pilot on privacy
  protected speaker identification

13
Agenda

• (Biometric) Identity Theft
• Privacy-by-Design
• privID for Private Speaker Identification
• Conclusion

14
SummaryPrivacy-by-Design solutions for
biometrics are pivotal for public acceptance of
biometrics

• Biometric is crucial to verify identities, at the
  risk of the perfect (biometric) identity theft
• Although at an early phase, Privacy-by-Design
  solutions promise intrinsic security
• Pseudo Biometric Identities
• Renewable and Revocable
• Additional benefits may drive down the cost of
  todays biometric systems
• Universal and Small Footprint
• Fast Search
• Universal Back-end for Multimodal

• European Initiatives
• 3D Face (FP6), Sagem, Philips. L1, Cognitec, FhG
  IGD
• New Opportunities
• Template protection for minutiae fingerprint for
  generating pseudo biometric identities

15
www.research.philips.com/password