The Inquirer-Home

Photonic chips to outperform conventional computers in five years

Quantum computing sees the light
Mon Jan 31 2011, 12:39

WITHIN FIVE YEARS quantum processors could be capable of simulations more complex than any that can be performed by the most powerful conventional computer today, according to the co-ordinator of a €2.2 million European project researching them.

Dr Mark Thompson of Bristol University's Centre for Quantum Photonics said that this is despite the fact that even the people designing the processors don't fully understand how they work. A general-purpose quantum computer capable of performing any calculation is decades away, but processors capable of specific tasks have already been demonstrated, he told a Cambridge forum on wireless research in January.

Researchers at Bristol are working on a project that would allow mobile phones to download cryptographic keys over a secure quantum photonic link. Each key can be used once only to secure a standard Internet transaction. "The idea is that your bank will give you a quantum chip for your mobile phone... and whenever you run out of keys you just point your phone at an ATM and download more," Thomson said.

He showed two Bristol-developed quantum photonic silicon chips, which depend on the movement of photons along waveguides rather than that of electronic charges along conductors like a conventional processor.

25534-originalOne of the chips performs a simple simulation and the other factors the number 15 as a proof of concept. The chips both require a source of single photons that are pumped in as input. Bristol recently announced a chip using two photons (photo credit: Jasmin Meinecke) and the computing power will increase vastly as the photonic count goes up. 

All of the chips depend on the little-understood phenomenon of superposition, whereby a photon can be in two places at once. Inside each is a mesh of waveguides designed to perform a particular quantum algorithm. When a photon reaches a fork it takes both paths at once, diffusing through the chip interfering with itself - or another photon if present - as it goes.

A photo-detector notes where each photon emerges from the chip and the result is inferred from the exit distribution.

Thompson, who co-ordinates the Quantum Integrated Photonics Circuits Euro project, said the technology has moved on from being equivalent to the valve and transistor stage of electronic computing and is producing its very first integrated circuits.

Thompson's particular interest is in using the technology for simulation at a time when even the most powerful conventional computer cannot model a molecule with 30 atoms.

One possibility is to figure out how photosynthesis works. "There are certain bacteria that can convert light to energy with 100 per cent efficiency and we want to know how they do it. Right inside the photosynthesis molecule there is a quantum effect going on. We are trying to develop quantum simulations that can model those effects," Thompson said.

Quantum computing presents both a threat and a promise. The technology can destroy financial life as we know it by providing tools to crack the encryption upon which it depends, but quantum computing also offers an alternative way to secure transactions. We wonder, is there a chance that the threat could become real before the solution is in place?

"We'd hope to see it coming in time... but we could roll out the solution pretty quickly if we had to," said Thompson. µ

Share this:

Comments
Busted!

Well exactly Zeitgeist Buster Buster, couldn't have put it better myself!

Wasn't it Dilbert who once said...

"Ignorance more frequently begets confidence than does knowledge"?

And while one has no faith in knowledge one does not feel confident enough to claim total ignorance.

posted by : Zeitgeist Buster, 01 February 2011 Complain about this comment
Bis repetita

Remember reading the same king of prediction when I started my PhD work on CuCl based optical memories in 1984 ;-)

posted by : MATHAE, 01 February 2011 Complain about this comment
Zeitgeist buster doesn't know if anyone knows and doesn't know he doesn't know

You don't know if they know how anything works, it is just a scientific conceit and a mathematical delusion that you think you know that they don't know.

It's an epistemological verity, but acting on the concept works, so you use it. Why not? It's what you think everyone else is doing anyway.

But if you think you know whether or not they know what is going on, you are kidding your self-referential argument.

posted by : Zeigeist Buster Buster, 01 February 2011 Complain about this comment
noone knows

how anything works anyway, it is just the scientist's conceit and the mathematician's delusion that they think they do

its an epistomological verity, but if acting on a concept works, use it, until it doesnt, why not? its what we are all doing anyway, but if you think you know what is going on, you are kidding yourself

posted by : Zeitgeist Buster, 31 January 2011 Complain about this comment
aboutus
Advertisement
Subscribe to INQ newsletters
Advertisement
INQ Poll

Facebook starts selling shares

Will you buy Facebook shares?