Contactless smart card | EPFL Graph Search

A contactless smart card is a contactless credential whose dimensions are credit-card size. Its embedded integrated circuits can store (and sometimes process) data and communicate with a terminal via NFC. Commonplace uses include transit tickets, bank cards and passports. There are two broad categories of contactless smart cards. Memory cards contain non-volatile memory storage components, and perhaps some specific security logic. Contactless smart cards contain read-only RFID called CSN (Card Serial Number) or UID, and a re-writeable smart card microchip that can be transcribed via radio waves. Overview A contactless smart card is characterized as follows: *Dimensions are normally credit card size. The ID-1 of ISO/IEC 7810 standard defines them as 85.60 × 53.98 × 0.76 mm (3.370 × 2.125 × 0.030 in). *Contains a security system with tamper-resistant properties (e.g. a secure cryptoprocessor, secure file system, human-readable features) and is capable of providing se

Biometrics & [and] Security

Claude Barral

Since the beginning of this brand new century, and especially since the 2001 Sept 11 events in the U.S, several biometric technologies are considered mature enough to be a new tool for security. Generally associated to a personal device for privacy protection, biometric references are stored in secured electronic devices such as smart cards, and systems are using cryptographic tools to communicate with the smart card and securely exchange biometric data. After a general introduction about biometrics, smart cards and cryptography, a second part will introduce our work with fake finger attacks on fingerprint sensors and tests done with different materials. The third part will present our approach for a lightweight fingerprint recognition algorithm for smart cards. The fourth part will detail security protocols used in different applications such as Personal Identity Verification cards. We will discuss our implementation such as the one we developed for the NIST to be used in PIV smart cards. Finally, a fifth part will address Cryptography-Biometrics interaction. We will highlight the antagonism between Cryptography – determinism, stable data – and Biometrics – statistical, error-prone –. Then we will present our application of challenge-response protocol to biometric data for easing the fingerprint recognition process.

EPFL2010

A Protection Scheme for MoC-Enabled Smart Cards

Claude Barral, Serge Vaudenay

The concept of Match-on-Card (MoC) consists of a smart card which receives an applicant's candidate template T to be compared with the stored reference template T_ref by processing the complete matching algorithm during a biometric authentication request. The smart card will then output whether this comparison is positive or not. The main argument against MoC-enabled smart cards is that it opens the way for YesCard (i.e. an attack path previously seen in Banking, a card always returning "yes"). The threat regarding Biometrics is not only YesCard, but also NoCard as we will see in this paper. We will propose a protocol to easily thwart these attacks by using simple cryptographic primitives such as symmetric encryption. This protocol will however only protect the system from malicious smart cards, but will not protect the smart card against malicious systems. Finally we will enhance this protocol to protect the smart card against its use as a so-called oracle to guess the stored reference biometric template.

IEEE2006

Implications of Position in Cryptography

Handan Kilinç Alper

In our daily lives, people or devices frequently need to learn their location for many reasons as some services depend on the absolute location or the proximity. The outcomes of positioning systems can have critical effects e.g., on military, emergency. Thus, the security of these systems is quite important. In this thesis, we concentrate on many security aspects of position in cryptography. The first part of this thesis focuses on the theory of distance bounding. A distance bounding protocol is a two-party authentication protocol between a prover and a verifier which considers the distance of the prover as a part of his/her credential. It aims to defeat threats by malicious provers who try to convince that they are closer to the verifier or adversaries which seek to impersonate a far-away prover. In this direction, we first study the optimal security bounds that a distance bounding protocol can achieve. We consider the optimal security bounds when we add some random delays in the distance computation and let the prover involve distance computation. Then, we focus on solving the efficiency problem of public-key distance bounding because the public-key cryptography requires much more computations than the symmetric-key cryptography. We construct two generic protocols (one without privacy, one with) which require fewer computations on the prover side compared to the existing protocols while keeping the highest security level. Then, we describe a new security model involving a tamper-resistant hardware. This model is called the secure hardware model (SHM). We define an all-in-one security model which covers all the threats of distance bounding and an appropriate privacy notion for SHM. The second part of this thesis is to fill the gap between the distance bounding and its real-world applications. We first consider contactless access control. We define an integrated security and privacy model for access control using distance bounding (DB) to defeat relay attacks. We show how a secure DB protocol can be converted to a secure contactless access control protocol. Regarding privacy (i.e., keeping anonymity in a strong sense to an active adversary), we show that the conversion does not always preserve privacy, but it is possible to study it on a case by case basis. Then, we consider contactless payment systems. We design an adversarial model and define formally the contactless payment security against malicious cards and malicious terminals. Accordingly, we design a contactless payment protocol and show its security in our security model. The last part of this thesis focuses on positioning. We consider two problems related to positioning systems: localization and proof of location. In localization, a user aims to find its position by using a wireless network. In proof of location, a user wants to prove his/her position e.g., to have access to a system or authorize itself. We first formally define the problem of localization and construct a formal security model. We describe algorithms and protocols for localization which are secure in our model. Proof of location has been considered formally by Chandran et al. in CRYPTO 2009 and it was proved that achieving security is not possible in the vanilla model. By integrating the localization and the secure hardware model, we obtain a model where we can achieve proof of location.

EPFL2018

Implications of Position in Cryptography

Handan Kilinç Alper

EPFL2018