Yubico Forum

<Preamble>
I haven't dug into RFID specs yet, but I have some prototype ideas bouncing around in my head.
Please excuse my ignorance in this area.

<Assumption>
My guess is that the RFID Yubikey will transmit an RF static value when exposed to the expected RF field.

<Question>
Do RFID readers transmit sufficient RF energy to induce a charge strong enough to increment the OTP?
If so, couldn't the RFID Yubikey be configured to transmit a OTP RFID response rather than a static one?

<Also>
If the induction was great enough (depending on how much time it might take to build sufficient charge and assuming that a capacitor could be added to the HW) could a reader also generate enough energy to power the capacitance touch dot to selectively determine if the YK should transmit a standard or a OTP response?


Thanks in advance for your feedback.
Cheers.


Craig.

Thanks for sharing your thoughts.

RFID for "contactless identification" typically works by the means of back-scattering. This means that the reader transmits enough RF power to supply the tag, thereby "reflecting" back the identity of the tag. The coupling is inductive and at 13.56 MHz, several milliwatts of power can be transmitted over to the tag, even with a small antenna so yes - a microprocessor can be powered by the field.

Our goal with the Yubikey RFID has not been to make the Yubikey contactless for normal authentication purposes as this would require a reader attached to the PC. The reason is rather to allow the Yubikey to be used for access control as well. Apart from the convenience aspect (one token instead of two), using the same authentication token for access to the PC as well as for physical access control fosters the behavior that users bring their tokens with them when they leave.

This means that we're somewhat limited to the established standards out there - making something home-cooked would require home-cooked readers as well. We've started with Mifare classic from NXP as it is widely spread. Mifare is not just about reflecting a static identity but a three-pass mutual authentication process. Although its crypto is broken, it is good enough for most access control purposes. We have other options on the menu depending on the interest.

With the best regards,

JakobE
Hardware- and firmware guy @ Yubico

JakobE,

Thank you for your response.
I appreciate the great deal of thought which has been applied to the development of the Yubikey and to its continued development.

You and your team are doing a great job.
...and thanks for the brief tutorial on RFID tech. I'm looking forward to researching this area.

Cheers,


Craig.