SoylentNews is people
SoylentNews
from the what-an-entangled-web-they-weave dept.
In the race for safer ciphers, China just quantum-leap frogged the rest of the world.
[...] Now, China aims to escape that contest entirely with the creation of a communication network not secured by math, but guaranteed by the fundamental rules of nature. A team has demonstrated mastery over the secret sauce behind such a "quantum internet" with their satellite Micius, which recently smashed the distance record for creating a bizarre link between light particles known as entanglement.
"They are years ahead of everyone else in this technology," says Vadim Makarov, head of a quantum hacking lab at the University of Waterloo in Canada, who was not involved. "It's absolutely awesome."
Launched August 2016, the Micius satellite successfully entangled photons between two Chinese towns almost 750 miles apart. The experiment bested former fiber-optics setups by a factor of 10, a feat chief architect Jian-Wei Pan says others dismissed as "a crazy idea" when he first proposed it back in 2003. The accomplishment proves possible the ultimate aim of cryptography: an invincible code system theoretically capable of instantly connecting any two (or more) points on Earth.
No Man-In-The-Middle for you!
(Score: 2) by driverless on Wednesday August 09 2017, @03:39AM (5 children)
That's not how "quantum cryptography" works. The very fact that the article talks about "safer ciphers" shows that the author doesn't understand it. It's a cool physics project, not anything to do with security. Every few years, someone posts a new record achieved, just like people announce being able to send a WiFi signal 200km using a dish the size of a small house on a mountaintop, or run/swim/cycle over X distance faster than anyone else. Congratulations China, have a biscuit, we'll wait for the next record in a year or two.
(Score: 2) by FatPhil on Wednesday August 09 2017, @07:56AM (4 children)
With mathematics, you can send out an ever-repeating message that, no matter how long you continue it, within the expected life of the planet, will never be decypherable. Nor will it be detectable that you've changed the payload to something else.
With quantum crypto, becuase an evesdropper is indistinguishable from noise, which is expected, eventually the ever-repeating message will yield enough bits to be decyphered. Similarly, a change to the payload will eventually be detected.
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 3, Interesting) by driverless on Wednesday August 09 2017, @09:17AM (3 children)
There's a lot more problems than just those. Since you can't observe the quantum-secured channel (without destroying the data on it), you can't make sure that it's working as required, which is a trivial operation for conventional crypto (pairwise consistency testing of the crypto is a standard requirement for FIPS certification). Another reason it's less safe is that you're now subject to a whole range of implementation glitches and flawed assumptions, some of which have already been noticed over time (see reports at arXiv [arxiv.org] and the IACR ePrint archive [iacr.org]), but many more of which are still to be discovered.
Even if you could overcome all of that, what you end up with is a very short-range version of unauthenticated Diffie-Hellman, circa 1976.
(Score: 2) by tangomargarine on Wednesday August 09 2017, @02:45PM (2 children)
Wouldn't that logic also preclude somebody from eavesdropping on it? Then the original recipient wouldn't receive the right data because it was observed?
I would think "but they could entangle the transmitter" would fall under "access to your hardware means you're fucked" unless you could do *that* long-distance, too.
"Is that really true?" "I just spent the last hour telling you to think for yourself! Didn't you hear anything I said?"
(Score: 0) by Anonymous Coward on Wednesday August 09 2017, @09:44PM (1 child)
Partially correct: the original recipient recipient will see bad data if it has been observed - but as the GP mentioned its indistinguishable from noise - so the recipient only knows he is either being observed, or there is noise on the line.
Assuming noise is a normal thing, everyone would always feel like they might be being observed...
(Score: 2) by tangomargarine on Thursday August 10 2017, @02:40PM
Assuming your communications provider is shit...yep, checks out :P
"Is that really true?" "I just spent the last hour telling you to think for yourself! Didn't you hear anything I said?"