[E-Lang] Quantum computing and capabilities

Ralph Hartley hartley@aic.nrl.navy.mil
Mon, 05 Feb 2001 14:04:14 -0500


Chris Hibbert wrote:

> Ralph Hartley wrote:
> 
>> But quantum cryptography is strictly stronger than one time pads,
>> because it does not require a trusted courier.
> 
> 
> QC requires something even stronger than a trusted courier: it requires
> an on-line connection between the parties.


Does it? In what sense? The distribution of the EPR pairs certainly 
doesn't need to be on line. The parties don't need to exchange pairs, 
they only need to obtain pairs from a common source. Because the source 
is not trusted, it can be part of a common infrastructure. If Bob wants 
to communicate with Alice he can obtain a bunch of Alice qbits from a 
public repository of bits that Alice has the mates to. Every thing else 
can be transmitted through adds in the Times. Of course he only knows 
that he is talking to whoever has the other half of the pairs, but the 
same thing is true of a public key.

In many of the protocols for quantum cryptography, the parties do 
interact on-line, but I don't know any reason that is required.

>   In exchange you get an
> assurance that the bits weren't tampered with, but I can imagine lots of
> circumstances when a trusted courier is feasible, and a fiber link (or
> the equivalent) is not.  One is Vinge's: interstellar communication.  Or
> am I missing something about setting up a quantum cryptographic 
> connection?


A qbit can be transmitted as a polarized photon. Fiber optics is a cost 
effective way to move photons around, but it's not the only way. They 
can also be transmitted trough empty space. The large mirrors needed to 
send and receive photons over interstellar distances are certainly MUCH 
cheaper than a star ship. You need to be able to send photons for 
unencrypted transmissions anyway. Quantum transmissions would favor 
short wavelengths, so that there are fewer high energy photons. Other 
considerations (bandwidth, antenna size) favor that too.

Theoretically you could send qbits through light bounced off a reflector 
on the moon or even HF radio bouncing off the ionosphere, but you would 
need very special transmitters/recievers, and it would be VERY inefficient.

Ralph Hartley