Re: New Light-Based Computer Runs At Quantum Speeds

From: scerir (scerir@libero.it)
Date: Fri May 18 2001 - 16:07:20 MDT


From: "Robert J. Bradbury"

> > New Light-Based Computer Runs At Quantum Speeds:
> > http://www.sciencedaily.com/releases/2001/05/010515075526.htm

> Does anyone know if this is any different from the work of
> Bucksbaum?

Yes, because they wrote:
" The caveat is that one should be careful of comparing
an inefficient classical algorithm with an efficient quantum one.
The experimental implementations of the Grover search to date
have not in fact implemented a search of an unsorted database,
but rather a database with one single marked item.
Therefore experimental evidence for the claimed speedup has
yet not been achieved, since a sorted database can be
searched classically in log 2 N steps, which is
identical to the readout limit for a quantum computer.
The superiority of the quantum search algorithm becomes
apparent only when one examines carefully the notion of a
query. It is evident that there is no information gain in the sort
of oracle query described above, which involves only unitary
transformations. It is in the encoding of such oracles, which
must be done by a quantum computer, that the real speedup
occurs. Once a properly encoded oracle is available, the number
of steps required to perform the information processing is
then limited by the final register readout to O(log 2 N).This
is exactly the same as for a classical information processor,
which shows that it is the realization of actual information
by the readout, as opposed to predictive information that is
contained in the quantum state, that is the ultimate limiting
procedure in quantum information processing.
We have shown here that if the readout of the register is
performed by particle counting then there exists a one-to-one
correspondence of single-particle quantum interference and
classical interference. Therefore we conclude that any enhancements
in processing power that can be ascribed to quantum
interference can also be found in classical wave processors,
and this includes systems based on modal entanglement.
Multi-particle entanglement, on the other hand, may provide
enhancements that cannot be efficiently transcribed to classical
interferometers, even when particle counting is used to
realize the output information. "

And then: ". As a consequence, computers
based on classical waves are as efficient as those
based on single quantum particles."

S. Wallentowitz, I.A. Walmsley, J.H. Eberly
http://arxiv.org/abs/quant-ph/0009069

"How big is a quantum computer?"

Accounting for resources is the central issue in computational
efficiency. We point out physical constraints implicit in
information readout that have been overlooked in classical
computing. The basic particle-counting mode of read-out sets a
lower bound on the resources needed to implement a quantum
computer. As a consequence, computers based on classical waves
are as efficient as those based on single quantum particles

Konrad Banaszek, Alfred B. U'Ren, Ian A. Walmsley
http://arxiv.org/abs/quant-ph/0103026

"Generation of correlated photons in controlled spatial modes
by down-conversion in nonlinear waveguides"

We report the observation of correlated photon pairs generated
by spontaneous parametric down-conversion in a quasi-phase
matched KTiOPO4 nonlinear waveguide. The highest ratio of
coincidence to single photon count rates observed in the 830 nm
wavelength region exceeds 18%. This makes nonlinear waveguides
a promising source of correlated photons for metrology and
quantum information processing applications. We also discuss
possibilities of controlling the spatial characteristics of the
down-converted photons produced in multimode waveguide structures.



This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:07:43 MST