I hadn't planed on writing on this subject so soon after my last post, but it
seems that lately I can't walk to my mailbox without finding another important
article about Quantum Computers. I keep thinking about Nuclear Physics in the
late 1930's.
Most people thought that error correcting codes for Quantum Information was
impossible, and this would be very bad news for Quantum Computers because
these machines will certainly make errors, no doubt about it. It turns out
however that most people were wrong, late last year Peter Shor of ATT found a
quantum error correcting code. The trouble was, although Shor's idea worked
well for storing and transmitting quantum information without error, it did
not work for the actual calculation, many thought that surely was impossible.
It turns out they were wrong about that too.
In the August 30 1996 issue of Science is an article by J. I. Cira,
T. Pellizzari, and P. Zoller entitled "Enforcing Coherent Evolution In
Dissipative Quantum Dynamics". They propose a Quantum error correcting scheme
with modest computational overhead that would dramatically increase the
number of quantum logic gates the machine could have before errors made it
unreliable. If p is probability that a single gate will fail, then without
error correction, a Quantum Computer can only have 1/p gates as a practical
matter. With this new quantum error correcting code it can have 4/p^2 gates
before errors overwhelm it. For example, if the probability that one gate
will fail is .09 then if you have no error correction your Quantum Computer
better not have more that 11 logic gates, with this new error correcting idea
it could have 494 logic gates without making more errors than the 11 did.
Apparently the appeal of making a calculation on 2^n numbers at the same time
with a machine that only has n qbits is too strong for the military to ignore.
In the same issue of Science is an article about the defense department
making a 5 million dollar grant to start an institute for Quantum Information
and Computing (QUIC). It's charter has 5 aims.
1) Improve quantum algorithms.
2) Improve quantum logic gates.
3) Improve the architecture of Quantum Computers.
4) Improve quantum error correcting codes.
5) Study the general theory behind quantum computation.
John K Clark johnkc@well.com
-----BEGIN PGP SIGNATURE-----
Version: 2.6.i
iQCzAgUBMjeWgX03wfSpid95AQHONQTwjuYh7RcgCGOG0Dnoge7fJ3qPKKAuxNni
pCH+a2k1X+s2NGyNOHbBU4OM+gtaOOx/3JznWqdzQAoWpmMi/3P15uJSNSzoepqz
QBKKPSKVwtQmecGqxcQCQHpw4mCMrDVHIZKnW3mhRtdhUrbLCWjOaG2edKuau92v
9jOs3UiFxfn49VbxUmiqohm1ZMpt+9cRde8aeOGj8fE2X/7H1gQ=
=Hl3S
-----END PGP SIGNATURE-----