Thread: RFID and Biometrics: A Bad Thing?

RFID and Biometrics: A Bad Thing?

Thursday, June 14, 2007 - RFID Connections


Bert Moore

Editor



Here's something to add to privacy concerns: some worry that the inclusion of biometric data on an RFID-enabled identification document actually increases the risk of identity theft than RFID alone. In fact, the concern applies to any method of encoding biometric data that can be read from a card without the individual's express permission. The other technologies identified as "security risks" were 2D symbols and magnetic stripes, both of which are cited as possible technologies for use on future US ID cards. The concerns were voiced in comments to the US Department of Homeland Security (DHS) in response to a call for comments on the proposed REAL ID Act.



The REAL ID Act is an attempt by the US Government to ensure that all state-issued IDs contain a basic minimum amount of information such as name, address, data of birth and a photograph and that these elements be encoded in "a common machine-readable format." There have been conflicting statements issued on whether magnetic stripe or 2D symbols would be used.



Regardless of the technology used -- RFID, magnetic stripe, optical symbols -- the question is whether inclusion of biometric data on a card does, in fact, pose a risk for identity theft.



In the case of fingerprint indicia, one of the most common forms of biometric identification, key features of a fingerprint are identified and stored in a database. A full fingerprint cannot be reproduced from this. However, the indicia itself -- the relative position of certain distinguishing characteristics of a fingerprint -- is unique which is why it can be used for identity verification.



The contention is that indicia can be "disassociated" from the ID card and used in unscrupulous ways. Even encryption is not adequate, according to some critics since any encryption algorithm would be static and, once broken, could provide no security. However, the specific ways in which this biometric data could be used are not elaborated.



If one makes only a cursory examination of the subject, one can imagine fingerprint indicia from a valid ID being copied onto a bogus ID to, essentially, clone an identity or hide a criminal one.



However, a closer examination of the subject would have to include the system in which biometric information would be used. Would it be compared only against a database -- in other words, validate the ID itself? Or would it be compared to the individual to verify identity?



If the purpose of the ID is to ensure that the holder of the ID is really who he or she says, then comparing the biometric data to the individual is the only valid approach. Skimming or cloning biometric information from a valid card would only highlight the fact that the card is bogus since it would not match the biometric data of the individual presenting it.



The concern about cloning biometric data assumes that there would be other uses for this data -- such as for access to buildings or financial transaction cards. This also assumes that criminals could hack into the biometric reader of the building's access control system or financial institution's ATM to feed in the cloned information in digital format.



Admittedly, if biometric data is used in the future to verify online financial transactions or enable remote access to a corporate computer, say, from a home computer, hackers could use cloned digital biometric data to masquerade as someone else.



This, however, assumes static private key encryption for biometric data. For magnetic stripe and 2D symbols where data capacity is limited by physical constraints, private key encryption may be all that's available. Advances in encryption technologies, however, make it clear that RFID tags will have more than adequate capacity to enable new public key encryption that will provide a much higher level of security than is currently being used in online financial transactions.



Since there is already great debate about the potential cost of issuing REAL ID cards, the additional cost of issuing smart cards would provide opponents of the REAL ID Act with more ammunition to kill or delay deployment of these cards which DHS sees as a necessary step in developing comprehensive homeland security systems.



While the expressed concern about the potential vulnerability of biometric data seemed to suggest that only smart cards are capable of providing adequate security, it also inadvertently makes the case for the use of advanced encryption technologies in less expensive RFID tags.



-----



Comments on this column? E-mail me: b.moore@aimglobal.org



http://www.aimglobal.org/members/news/t ... &zoneid=24

Thanks, JP! Great article!

You are welcome Rockfish!