Quantum scheme lightens load

By Eric Smalley, Technology Research News

Two years ago, scientists proved it possible to build a quantum computer from simple optical equipment commonly found in university classrooms and laboratories. Now researchers at Johns Hopkins University have refined the approach, reducing the amount of equipment linear optical quantum computers would need by about two orders of magnitude.

Quantum computers use the weird nature of particles like atoms, electrons and photons to perform many computations in parallel. If a big enough quantum computer could be built, it would far outstrip classical computers for solving certain problems like cracking secret codes. So far, however, only the most rudimentary quantum prototypes have been constructed.

The Johns Hopkins plan shows that equipment like mirrors, half mirrors and phase shifters could be used to make practical, photon-based quantum computers, said James Franson, principal staff at the Johns Hopkins University Applied Physics Laboratory and a research professor the university's electrical and computer engineering department. "Our approach may make it more feasible to develop a full-scale quantum computer," he said.

Controlling single photons using linear optics equipment is simpler than manipulating individual or small numbers of atoms or electrons, which are the basic units of most other quantum computing schemes.

Capturing and manipulating atoms and electrons involves precisely tuned lasers or magnetic fields, or carefully constructed microscopic devices. It's also much harder to transport isolated atoms and electrons than it is to move photons. "An optical approach to quantum computing would have a number of potential advantages, including the ability to connect different devices using optical fibers in analogy with the wires of a conventional computer," said Franson.

Linear optical quantum computers, like ordinary electronic computers, would use circuits that link simple logic devices in intricate patterns that make the output from one device the input to the next. The 1s and 0s of linear optical quantum computing would be represented by properties of photons like horizontal versus vertical polarization rather than the presence or absence of a current of electrons.

The potential power of any type of quantum computer comes from its ability to examine all possible solutions to a problem at once rather than having to check one at a time.

This is possible because when a particle like a photon is isolated from its environment it is in the weird quantum state of superposition, meaning it can be horizontally and vertically polarized at once, and so can represent a mix of 1 and 0. This allows a string of photons in superposition to represent every combination of 1s and 0s at the same time so that a quantum computer could process all the numbers that represent possible solutions to a problem using one set of operations on the single string of photons.

Linear optical devices perform quantum logic operations by altering photons according to probabilities. Half mirrors, or beam splitters, for example, can direct photons along one of two paths, with an even chance for each path.

The challenge of linear optical quantum computing is to pass the correct result of a quantum logic operation from one device to the next without directly observing the states of the photons that represent the results, because this would change the states and therefore destroy the information the photons contain.

The trick is to put additional photons through the logic operation at the same time. These additional, ancilla photons trigger the optical circuitry that passes along the output of the logic operation when the result of the operation is correct. The ancilla photons are absorbed in photon detectors in the circuitry, but the output photons are preserved and passed on.

The key advance in the Johns Hopkins researchers' approach is that it uses fewer ancilla photons by entangling input and ancilla photons in a way that minimizes the probability of errors, said Franson. When two or more particles in superposition come into contact with each other, they can become entangled, meaning one or more of their properties change in lockstep even if the particles are separated.

Fewer ancilla photons means fewer pieces of equipment are needed. "Using the current error correction techniques, our high-fidelity approach should reduce the [equipment] required by roughly two orders of magnitude," said Franson. The amount of equipment required to generate the entangled ancilla state and the probability of an error "both increase rapidly with increasing numbers of ancilla photons," he said.

The original linear optical quantum computing scheme had an average error rate of 2/n, while the researchers' refined scheme has an average error rate of 4/n2, according to Franson. N represents the number of ancilla photons. This translates to error rates of 20 percent versus 4 percent for 10 ancilla photons, and 2 percent versus 0.04 percent for 100 ancilla photons.

This gives the Johns Hopkins scheme a practical error rate with far fewer ancilla photons, said Franson. Quantum error correction will require error rates on the order of 0.1 to 0.01 percent, he said. "That range of errors could be achieved with 100 ancilla in our case, but that would require 5,000 ancilla in the original... method."

Because the scheme requires fewer mirrors and beam splitters to manipulate the smaller number of ancilla photons, it makes it more likely that a practical linear optical quantum computer could be built, said Jonathan Dowling, supervisor of the quantum computing technologies group at NASA's Jet Propulsion Laboratory. The researchers' method "seems to be a substantial improvement over the original scheme," he said.

Devices enabled by this new approach will be used in quantum communications systems before they are used in full-blown quantum computers, said Dowling. With experience gained from making quantum communications devices, the researchers' approach will eventually lead to "a practical, compact, all-optical quantum computer," he said.

Dowling's group has developed a plan for a quantum repeater, a device necessary to boost quantum communications over long distances, that is based in part on the researchers' linear optical quantum logic, said Dowling.

The researchers have shown that the overhead needed to achieve a given fidelity for linear optical quantum logic gates can be significantly improved, said Emanuel Knill, a mathematician at Los Alamos National Laboratory and one of the scientists who developed the concept of linear optical quantum computing.

The Johns Hopkins researchers' approach does not address logical qubits, however, said Knill. Logical qubits are encoded from two or more physical qubits, and this makes them more resistant to errors. "My preference is to use logical qubits," said Knill. "If one wishes to use physical, not logical, qubits, then the authors' approach would help significantly," he said.

Quantum repeaters could be developed in five years, said Franson. "Full-scale quantum computers would be much more difficult and would probably require 15 to 20 years in the most optimistic scenario," he said.

The researchers are working on making photon-based logic gates and memory devices, and single-photon sources, said Franson. "These are the basic building blocks of a linear optics approach to quantum computing," he said.

Franson's research colleagues were Michelle Donegan, Michael Fitch, Bryan Jacobs, and Todd Pittman. They published the research in the September 23, 2002 issue of the journal Physical Review Letters. The research was funded by the Office of Naval Research (ONR), the Army Research Office (ARO), the National Security Agency (NSA) and the Department of Defense (DOD) Independent Research and Development Program (IR&D).

Timeline:   5 years, 15-20 years
Funding:   Government
TRN Categories:   Physics Quantum Computing and Communications
Story Type:   News
Related Elements:  Technical paper, "High-Fidelity Quantum Logic Operations Using Linear Optical Elements," Physical Review Letters, September 23, 2002


October 16/23, 2002

Page One

Chemists brew tiny wires

Voiceprints make crypto keys

Stamp corrals tiny bits

Net devices arranged fractally

Quantum scheme lightens load


Research News Roundup
Research Watch blog

View from the High Ground Q&A
How It Works

RSS Feeds:
News  | Blog  | Books 

Ad links:
Buy an ad link


Ad links: Clear History

Buy an ad link

Home     Archive     Resources    Feeds     Offline Publications     Glossary
TRN Finder     Research Dir.    Events Dir.      Researchers     Bookshelf
   Contribute      Under Development     T-shirts etc.     Classifieds
Forum    Comments    Feedback     About TRN

© Copyright Technology Research News, LLC 2000-2006. All rights reserved.