A better way to handle encrypted data

0 14

Te digitization of modern life means that data security is ever more important. Data in storage and transfer is usually encrypted, thus safe from prying eyes. But to work, they usually need to be unencrypted first. This is a particular problem with so-called cloud computing (actually, just a row upon a row of stacks of computers in server farms), which happens outside the control of the data owner. And it’s getting worse, as more and more devices report back to different clouds, rather than doing them locally.

Listen to this story.
Enjoy more audio and podcasts ahead iOS or Android.

Your browser does not support the element

Save time by listening to our audio articles while you multitask

A possible answer is a way in which encryption is fully homomorphic (FHE). This allows direct computation of encrypted data. Someone with the right key, could use FHEsend information to a cloud, have it processed there, and get the results back without putting anything sensitive at risk.

The problem with this approach is that it is slow. Very slow. Nick New, head of Optalysys, a small British company, says that a calculation that takes one second on unencrypted data might take 1m seconds with FHE. Mr New’s answer, as his company’s name suggests, is to use optical rather than electronic computing.

Optical computing encodes data in light beams instead of electrical currents. The calculation is done by manipulating the beams. This works well for a type of math called linear algebra – and fortunately, Fourier transforms, a way to speed up all the multiplications involved. FHE, easy to handle this way. Mr. New thinks FHE and optical computing together would reduce the processing time of a mass problem from 1m seconds to between ten and 100.

In the Optalysys system, the information is encoded into the level and amplitude of many different behaviors. These are then directed in specific directions by structures called waveguides, before being radiated into free space, where they combine and interfere with each other.

The calculation takes place when the resulting wave front is passed through a specially designed lens, and the result is translated into an electrical signal, for more conventional processing, by a camera. An advantage is that, as with all optical computing, the actual calculation does not consume energy, saving money and carbon dioxide emissions.

Do you know about the world? To enjoy our mind-expanding science coverage, sign up to Simply Science, our weekly subscription-only newsletter.

Leave A Reply

Your email address will not be published.