Magazine article Science News

New Method Creates Real Randomness: 'Extractor' Removes Bias from Computer-Generated Numbers

Magazine article Science News

New Method Creates Real Randomness: 'Extractor' Removes Bias from Computer-Generated Numbers

Article excerpt

Ask a computer to pick a random number and you'll probably get a response that isn't completely unpredictable. Because they are deterministic automatons, computers struggle to generate numbers that are truly random. But a new advance on a method known as a randomness extractor makes it easier for machines to roll the dice, generating truly random numbers by harvesting randomness from the environment.

The method improves on previous randomness extractors because it requires only two sources of randomness, and those sources can be very weak. "It's a big breakthrough on a fundamental problem," says computer scientist Dana Moshkovitz of MIT.

Eshan Chattopadhyay and David Zuckerman, computer scientists at the University of Texas at Austin, were scheduled to present the new randomness extractor June 20 in Cambridge, Mass., at the Symposium on the Theory of Computing.

For computers, random numbers are a precious resource, essential for encrypting sensitive information such as credit card numbers, for instance. But computers typically fail at generating truly random numbers. Many computer applications instead rely on "pseudorandom" numbers. These are generated in a reproducible way, relying on an algorithm, and therefore aren't really random.

Deviations from true randomness can create security holes. "A common way for hackers to break into systems is to exploit the fact that people don't use high-quality randomness," says Zuckerman.

To sidestep computers' predictable natures, computer scientists have devised ways of harvesting randomness from the environment, using input from the mouse or the keyboard, for example. The computer might sample the mouse's coordinates at several points in time and convert these values into a string of numbers. But this still falls short of being truly random. If the mouse is on the left of the screen one moment, it's less likely to be all the way on the right in the following instant. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.