Quantum computers should be much easier to build than previously thought, because they can still work with a large number of faulty or even missing components, according to a study published in Physical Review Letters.
"Quantum computers can exploit this weirdness to perform powerful calculations, and in theory, they could be designed to break public key encryption or simulate complex systems much faster than conventional computers," said Dr Sean Barrett, the lead author of the study, who is a Royal Society University Research Fellow in the Department of Physics at Imperial College London.
The machines have been notoriously hard to build, however, and were thought to be very fragile to errors. In spite of considerable buzz in the field in the last 20 years, useful quantum computers remain elusive.
Barrett and his colleague Dr. Thomas Stace, from the University of Queensland in Brisbane, Australia, have now found a way to correct for a particular sort of error, in which the qubits are lost from the computer altogether. They used a system of 'error-correcting' code, which involved looking at the context provided by the remaining qubits to decipher the missing information correctly.
The findings indicate that quantum computers may be much easier to build than previously thought, but as the results are still based on theoretical calculations, the next step is to actually demonstrate these ideas in the lab. Scientists will need to devise a way for scaling the computers to a sufficiently large number of qubits to be viable, says Barrett. At the moment the biggest quantum computers scientists have built are limited to just two or three qubits.
References
Sean D. Barrett, Thomas M. Stace. Fault tolerant quantum computation with very high threshold for loss errors. Physical Review Letters, 2010
Imperial College London (2010, November 13). Quantum computers easier to build: Can tolerate faulty or missing components, researchers say. ScienceDaily. Retrieved November 13, 2010, from http://www.sciencedaily.com/releases/2010/11/101109081606.htm
No comments:
Post a Comment