Using virtual particles to get real random numbers
Bouncing photons off ‘empty space’
number generator, by bouncing photons off empty space.
As regular readers know, empty space isn’t actually empty – it’s home to “quantum fluctuations”, one of the stranger manifestations of quantum theory, in which particles and anti-particles spontaneously come into existence and annihilate so as to preserve the uncertainty principle. And if that’s not strange enough for you, why not learn how to turn virtual photons into real ones?
The new trick, demonstrated by a group led by Benjamin Sussman of Canada’s National Research Council in collaboration with researchers at the NRC and Oxford University, exploits vacuum fluctuations to create random numbers.
A big problem with “random” numbers generated by computers is that while “true” randomness is really useful for cryptography, it’s also very hard to achieve. Even very good computer algorithms are deterministic in some way. Quantum systems, on the other hand, are non-deterministic – or, to be more accurate, individual measurements might not be. As the researchers’ paper states:
“While the evolution of a quantum function is deterministic, the outcome of a particular measurement on a state is not”. If the right observable characteristic of a quantum system is chosen, the paper argues, a “non-deterministic” – and therefore truly random – number can be achieved.
This is already known and used in quantum random number generators, with the paper citing techniques such as vacuum shot noise (measuring the quantum noise in a signal), radioactive decay, laser noise, photon statistics, or fluorescence from entangled ions. However, many of these are too slow to keep up with the requirements of data networks.
Sussman’s technique is essentially to bounce light pulses off the “virtual particles” of quantum fluctuations, which randomizes the signature of the received light, and, because of the characteristics of the system he’s using, happens very quickly (Sussman claims Gbps key generation is possible, although his experiment is limited by the instruments available to 1 kHz).
Here’s how it works: a light pulse (“pump” pulse) is shone into a 3mm diamond plate, which generates a Stokes field with random phase. The original pulse is filtered out, leaving only the Stokes field, which is then combined with a reference pulse.
It’s what happens inside the diamond, however, that’s interesting: the interaction of the pump beam with the vacuum fluctuations changes the phase of the incoming photons in picoseconds, the paper claims, and since phase can be very precisely measured, Sussman says it’s possible to generate multiple bits per measurement. ®
Sponsored: Hyper-scale data management