Classical technique may outperform quantum cryptography

Quantum cryptography may be theoretically secure, but in practice, there are certain limitations that allow clever attackers to read encrypted messages. A new system that ditches quantum mechanics for classical mechanics may prove to be even more secure, backed up by the Second Law of Thermodynamics.

The advantage of quantum cryptography lies in quantum randomness, but so do the problems: since it's impossible to transmit a message without introducing errors, an attacker can use those errors to intercept parts of that message without anyone knowing, and there's no way around that. Sometimes, weird quantum randomness is just weird and quantum and random, and it's possible that a simple electrical connection might actually offer better security.

Here's how this new system, developed by researchers at Texas A&M University, works: you've got Alice, your secret agent, who needs to transmit a message to Bob, her handler back in the mountaintop underwater volcano lair. Meanwhile, Eve (who's Eve-il) is going to try and listen in. Alice and Bob hook up to the same bit of wire (a very long bit of wire), and Bob uses a generator to send a random signal through said wire. On her end, Alice has two different resistors, which are electronic thingies that can alter the current in a circuit, and Bob has the same thing on his end too. One of these resistors corresponds to a zero bit of data, and the other one corresponds to a one bit.

When Alice wants to send a message to Bob, she hooks up the appropriate resistor (the zero resistor or the one resistor, depending on what bit of data she wants to send) to the circuit. At the same time, Bob picks one of his resistors at random and hooks it up too. The combination of resistors changes the current in the wire, and since Bob knows how much current his generator is putting in as well as how his random resistor is changing things, he can solve the system for what resistor Alice must have used on her end to figure out what bit of data she's sending.

Meanwhile, let's check in with Eve, who's busy tapping the wire in between Alice and Bob. As far as Eve can tell, the voltage and current flowing through the wire is totally random, thanks to the random signal that Bob's generator is putting out. When Alice transmits her data, the current will change a little bit, but here's the clever part: the change in current also depends on the resistor that Bob picked at random, and as long as Bob keeps on making perfectly random choices, there's absolutely no way for Eve to work out which resistor Alice is using. Furthermore, if Eve tries to actively get more information out of the system by sending any signals of her own, Alice and Bob will be able to immediately tell, since the current that they're measuring will be altered.

The authors of this technique say that what makes their method so secure against passive listening is the Second Law of Thermodynamics, which states that the overall entropy of a system must always increase, and therefore their technique will become vulnerable shortly after the invention of a perpetual motion machine. We'll level with you: we've read the paper and done some research and still aren't entirely able to explain why the Second Law is responsible for the security, but even without that understanding, it still seems like this is a fairly impenetrable new system.

Check it out for yourself (and feel free to try and explain it to us!) by reading the full paper at the link below.

arXiv, via MIT

(Thanks, Stephen!)

For the latest tech stories, follow DVICE on Twitter
at @dvice or find us on Facebook