Completely secure networks?

[Vorlin]

Nothing will ever be secure. See the Great Wall of China, which was soon circumvented by the bribery of the guards.

Anything with human intervention/interaction in it has a chance for failure, buyout, bribery, espionage, corruption, infiltration, and error.

[foxyloxley]

As far as I can make out, there isn't any data attached to the photon. It is the sequence of photons that make the message.

And to intercept the photon isn't a runner either. The photon is intercepted by colliding with another photon, this action releases energy, which is recordable, and gives the position of THAT photon THEN. The photon has now been deflected to a different path, and will not arrive in time or place. The interceptor will be aware that there WAS a photon, and they will know where it WAS (both past tense)

And remember that the probability is that this protocol would more than likely be used over real secure systems that no one is going to tamper with. example, the fibre optic links that are used to transmit National Security details. So physical security is of a high standard to begin with.

The real trick here is to remember what a photon is: it is a piece of LIGHT. it has physical properties, mass, dimension etc, but it is damn small measurment in anyones money.......

[Tim_axe]

I guess that would make it a reasonably well thoughout and implemented way to transfer keys then. I don't think many people are going to be tapping into busy corporate fiber networks and risk exposing themselves during the downtime the network faces while they work on splicing everything together or something. I was just trying to get that feeling of "what if" out, and you've been a good help. Thanks

[foxyloxley]

I wouldn't want to make people think that this is a speciality of mine...........

I got into the Quantum thing after reading about [amongst others] Alan Turing and Stephen Hawking. From these I started to read the books by Marcus Chown and my particular favourite, John Gribben.

The books by J Gribben have started me off on that voyage of discovery, where I am quickly in over my head, and these are books MEANT for Joe Public...............

They attempt to explain the inexplicable, they leave you with more questions than when you started, and all this ON TOP of reading up on the basics in networks and systems AND day to day developments in the IT world.

And, of course, coming here to AO almost every day as well........

What I'm TRYING to say is: Don't let me be the one to stop you asking questions, just be prepared to get MORE questions along with the occasional answer.

My 2c

[djscribble]

the other thing that you have to keep in mind is that security is like a teeter-totter where security is on one end and useablity is on the other. the more secure soemthing is the less useable. you want a secure network?? sure, don't plug it in, don't have internet access, forget the passwords, don't have physical access, have it protected by serin gas, etc... as you can see the most secure network is one that cannot be used

[hellforgedangel]

First off the idea of Quantum Cryptography is rather old (a few years). Secondly a major floor in the system is that by intercepting you cause the legitimate users to lose the data. This is a big problem on a network thats worth protecting in the first place, Think how easy a DOS attack would be if all you needed to do was place a sniffer on the line and there would be 100% packet loss.

[foxyloxley]

for hellforgedangel:
To DOS a quantum system...............

The arguments are about a SINGLE photon..........

to put it to 'perspective' A SINGLE volt has BILLIONS of electrons passing a fixed point every second, And electrons can produce photons.

http://www.free-definition.com/Photon.html

So the photon is a very very very small bit. It also travels at the speed of LIGHT........

So, we have a carrier voltage of say 30v, = approx 1 TRILLION electrons per second, ANY of these can be forced to produce the photon, AND it's all happening at SOL1 [Star Trek anyone] And onto this you are going to introduce a sniffer to screw up the system.

A sniffer is used to pick up details of a PACKET. relative size to photon = F**king HUGE.

It is this question of scale that will give the quantum system it's security, AND its Achilles Heel.

The maths to make this happen is being developed now, it will have its problems, but it WILL work eventually. For a given period of eventually, obviously

As said earlier, it will be used on VERY secure systems at first, but as in all things it will filter down to the man in the street. And on that day, you can kiss goodbye to PGP et al.................

another 2c

[nihil]

OK Foxy, et al................at the end of the day we have two "decryptors"........wel the sender and the recipient?

Grab one of those, and you are in, no matter how clever your encryption or transmission protocols are?

That's how I see it.............not that different from normal computer security.........allow physical access and you are doomed.

[Spiritblades]

Very interesting article..very interesting. I can't think of any way to compromise this system, other than having physical access to both the systems and the right tools to do the job, and performing some sort of a middleman deal with this..that would get you the key and all the data along w/it..but $.$