Conventional
encryption generates significantly more data than Algebraic Eraser, says Parks, which grows exponentially as users dial up the level of encryption desired by using longer keys. Generally, he explains, the more data you need to encrypt or decrypt, the more processing power and time your system requires.
Juels notes that on paper, what SecureRF is proposing sounds highly implausible. However, he knows the mathematicians involved in the development of the Algebraic Eraser are well known and respected. Whether it will work, he says, won't be known until the
protocol is built into tags and tested.
"One of the fundamental laws of cryptography is that a system is not to be trusted without peer review and focused scrutiny," he says. "This is particularly important with
RFID tags that use cryptography, because they're very hard to patch. If there is a flaw in a browser, or Microsoft finds a flaw in its operating system, it deploys a patch over the Internet. You can't do that when you have a billion little hardware devices floating around and you find a flaw [in the encryption]. I would hope that Secure RF will publish its algorithms and provide time for their review."
Parks says SecureRF is working with a number of cryptographers to perform a peer review and search for weaknesses in the Algebraic Eraser. The algorithm it uses is patent-pending. "SecureRF is a little over a year old, but its roots go back over a decade. The mathematician-cryptographers who are responsible for this breakthrough have been working in the cryptography field for more than a decade and have invented the genesis of this new security platform," says Parks.
SecureRF is planning initially to market the Algebraic Eraser to RFID and smart card applications. However, Parks says the product is well-suited for any applications requiring fast and secure data encryption, such as WiMax (802.16), an emerging standard for high-bandwidth wireless networks. It could also be used, he says, to secure data sent via satellite communications and cell phones.