Hardware RFID Security for Preventing Far-Field Attacks

Radio frequency identification (RFID) is a widely used approach for a short-range contactless data exchange, i.e., employed in billing systems. During the last years, unauthorized access has become an issue, as it has been proven on numerous unpleasant occasions. A typical theft scheme is based on approaching a victim with a card reader. However, a straightforward adaptation of the reader’s antenna elements allows performing the attack from a distance, opening a severe security loophole. Here, we propose and demonstrate hardware-based protection capable of preventing a far-field attack of this kind. Our solution is based on an RFID chip shielding with an opaque metal. The Faraday type of the enclosure has a small aperture, which suppresses electromagnetic field leakage from the device. This property affects both up and down interrogation links, virtually making the far-field attack impossible. Activation of the card is done by holding it in hand–this way, it is made accessible to an authorized readout, which still remains wireless. The physical principle of the operation is placing a high-index dielectric structure next to a subwavelength aperture, making it electromagnetically larger and, as a result, supporting the field leakage. Our experimental prototype, validated by several different users, shows the capability to diminish far-field attacks on ultrahigh frequency (UHF) RFID tags. This hardware security solution can find usage in numerous applications, such as biometric passports, credit cards, and many others, where unauthorized access to sensitive data is highly undesirable.

I. Introduction

Wireless communication security has become an important topic over the past years owing to numerous thefts of different kinds. Data, being transmitted in a classical public channel, can be collected by a third party, causing a big privacy issue. Numerous security protocols have been developed to address those issues and allowed using cellular communications and other wireless applications rather safely [1]. However, several fundamental loopholes remain and seem to be unsolvable with software encryption approaches. A celebrated example is radio frequency identification (RFID), where data are encoded on a chip connected to an antenna [2]. This passive device is interrogated by an active reader. In this communication protocol, data are encoded on time-modulated back reflection from a tag. Being standardized, this wireless channel is subject to theft since any card obeying the standard should be accessible by any reader. A typical example here is data theft from contactless credit cards [3], [4]. One of the theft schemes is based on approaching a victim with a billing terminal and charging a small amount, which does not require a PIN-code authorization. While nowadays this theft scheme is applied in crowded places, where physical proximity with a victim is less alarming, RFID theft from a range is also possible [5], [6]. It is worth noting that credit cards operate at 13.6 MHz, though our investigations will concentrate on ultrahigh frequency (UHF) RFID, operating around 900 MHz, depending on a licensing country [7].