Turbo-charged quantum crypto? You'll need Cambridge laser boffins for that
Pushing one laser beam inside another
Boffins hope to turbo-charge the speed of “unbreakable” quantum cryptographic systems with a new technique involving “seeding” one laser beam inside another.
Researchers from the University of Cambridge and Toshiba Research Europe have used the technique to demonstrate that it might be possible to distribute encryption keys at rates between two and six orders of magnitude higher than was previously possible with real-world quantum cryptography systems.
The results are reported in full in the journal Nature Photonics (link here, subscription required for access to full article).
Quantum cryptography promises “unbreakable” security by harnessing the weird and wonderful world of quantum mechanics. Keys to encoded messages are transmitted by a sender using polarised photons. The recipient of the messages uses photon detectors to measure which direction the photons are polarised in, which signify whether a bit is either a one or a zero.
Thanks to of the laws of quantum cryptography, any attempt to eavesdrop on the message in transit would change its contents, allowing parties in a quantum key-enabled key exchange to detect eavesdropping and drop compromised keys.
That’s the theory. In practice, problems with quantum cryptography arise when trying to construct a usable system. It is a back-and-forth game: inventive attacks targeting different components of the system are constantly being developed, and countermeasures to foil attacks are constantly being developed in response.
The University of Cambridge and Toshiba Research Europe team have developed a new quantum cryptography protocol, based on different encoding, that is both faster and (it’s hoped) more secure, as the researchers explain. Alice and Bob are the sender and recipient of the encrypted messages.
The components that are most frequently attacked by hackers are the photon detectors, due to their high sensitivity and complex design – it is usually the most complex components that are the most vulnerable. As a response to attacks on the detectors, researchers developed a new quantum cryptography protocol known as measurement-device-independent quantum key distribution (MDI-QKD).
In this method, instead of each having a detector, Alice and Bob send their photons to a central node, referred to as Charlie. Charlie lets the photons pass through a beam splitter and measures them. The results can disclose the correlation between the bits, but not disclose their values, which remain secret. In this set-up, even if Charlie tries to cheat, the information will remain secure.
MDI-QKD has been experimentally demonstrated, but the rates at which information can be sent are too slow for real-world application, mostly due to the difficulty in creating indistinguishable particles from different lasers. To make it work, the laser pulses sent through Charlie’s beam splitter need to be (relatively) long, restricting rates to a few hundred bits per second (bps) or less.
The method developed by the Cambridge researchers gets around this limitation by using a technique known as pulsed laser seeding, in which one laser beam injects photons into another. This makes the laser pulses more visible to Charlie by reducing the amount of “time jitter” in the pulses, so that much shorter pulses can be used. Pulsed laser seeding is also able to randomly change the phase of the laser beam at very high rates.
The practical upshot is that pulsed laser seeding can generate data transfer rates as high as 1 megabit per second, representing an improvement of two to six orders of magnitude over previous efforts. We are one step closer to being able to use the MDI-QKD technique in the real world, thanks to the techniques pioneered by the University of Cambridge and Toshiba Research Europe.
"This protocol gives us the highest possible degree of security at very high clock rates," said Lucian Comandar, a PhD student at Cambridge’s Department of Engineering and Toshiba’s Cambridge Research Laboratory. "It could point the way to a practical implementation of quantum cryptography." ®