Biometric template protection in multimodal authentication systems based on error correcting codes^*

Issue title: EU-Funded ICT Research on Trust and Security

Guest editors: Jan CamenischGuest-Editor, Javier LopezGuest-Editor, Fabio MassacciGuest-Editor, Massimo CiscatoGuest-Editor and Thomas SkordasGuest-Editor

Article type: Research Article

Authors: Argyropoulos, Savvas^{a; b} | Tzovaras, Dimitrios^{a; **} | Ioannidis, Dimosthenis^a | Damousis, Yannis^a | Strintzis, Michael G.^{a; b} | Braun, Martin^c | Boverie, Serge^c

Affiliations: [a] Informatics and Telematics Institute, Centre for Research and Technology Hellas, Thermi-Thessaloniki, Greece | [b] Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, Thessaloniki, Greece | [c] Continental Automotive, GmbH, Regensburg, Germany

Correspondence: [**] Corresponding author: Dimitrios Tzovaras, Informatics and Telematics Institute, Centre for Research and Technology Hellas, Thermi-Thessaloniki, GR 57001, Greece. Tel.: +302310464160; Fax: +302310464164; E-mail: dimitrios.tzovaras@iti.gr.

Note: [*] This work was supported by the European Commission under contract FP6-026990 HUMABIO.

Abstract: The widespread deployment of biometric systems has raised public concern about security and privacy of personal data. In this paper, we present a novel framework for biometric template security in multimodal biometric authentication systems based on error correcting codes. Biometric recognition is formulated as a channel coding problem with noisy side information at the decoder based on distributed source coding principles. It is shown that the proposed method binds the biometric template in a cryptographic key which does not reveal any information about the original biometric data even if it is compromised by an attacker. Furthermore, the advantages of the proposed method in terms of security and impact on matching accuracy are discussed. We assess the performance of the proposed method in the context of HUMABIO, an EU Specific Targeted Research Project, where face and gait biometrics are employed in an unobtrusive application scenario for human authentication. Experimental evaluation on a multimodal biometric database demonstrates the validity of the proposed method.

Keywords: Multimodal biometrics, authentication, biometric template security, distributed source coding

DOI: 10.3233/JCS-2010-0369

Journal: Journal of Computer Security, vol. 18, no. 1, pp. 161-185, 2010