MIT and Texas Intruments create an RFID Chip that is secure.. for now

The term hack-proof is one that people have thrown around for some time. It is one of those phrases like NSA-Proof or malware proof that really do not mean what they should mean. However, it is certainly meant to mean that the technology in question is resistant to the majority of known (important word there) attacks. It is important to mention this definition as we start talking about the announcement that MIT and Texas Instruments have developed a new “hack-proof” RFID chip.

The announcement is on MIT’s new page and talks about the discovery and starts off with a line that is sure to grab your attention: “virtually impossible to hack”. Reading further the concept is that the envisioned technology primarily protects against side-channel attacks. This type of attack tries to identify patterns in the exchange of information between the RFID chip and reader. These patterns can, over time, allow someone to steal the secret key for the chip. Once you have this, you can compromise the system. There are several ways that an attacker can try to gain access to the secret key depending on how the key is implemented. If it is a stock key (issued at programing) then they look for leakage during multiple uses to learn the key. In systems that rotate keys through a random number function then a killing the power to the device during a read (over and over again) can yield results. You simply cut the power to the chip before the system can reset the encryption key and you continue to work against the same key.

The MIT researchers came up with a couple of designs to offset this limitation. One was to add a power supply to the RFID chip so that cutting the power would be very difficult to cut or to interrupt. It does not mean that it is impossible, just very, very hard. Another idea is to use non-volatile memory on the chip so that the data transferred to the chip and will be held in the event of a power loss. Now, adding memory or power to a credit card might not be the easiest thing to do as it requires a lot of space.

Fortunately the same exact material might be used to accomplish both of these requirements. Called ferroelectric crystal it can act like a capacitor and also maintain information on it after a loss of power. According to the MIT research Texas Instrument has been using this type of material to put non-volatile memory into systems for a while. When acting as NVRAM the crystals react to the application of electricity: “The application of an electric field, however, can align the cells’ polarization in either of two directions, which can represent the two possible values of a bit of information.”

Now, here is the cool part. Because the crystals are capable of retaining the information they can also retain a charge. Ti has built them in both 1.5 and 3.3 volt flavors. The new chip would have a group of 3.3v cells that work in combination with the 1.5v cells. When the chip hits the reader it would charge the 3.3 volt cells while it writes data to the 1.5v cells. If the power is interrupted during that time the 3.3 volt cells will keep power while the chip completed the task of updating the encryption keys or any other tasks it was working on.

If these chips can be produced in mass volumes and inexpensively they would certainly alleviate one of the main attacks on RFID systems. However, these are only the current batch of problems that are out there. Threat actors are very clever and will quickly adapt to the new technology. In fact we are positive that they are already working on something now that this news is out.

Check out the details

No comments

Leave your comment

In reply to Some User