Solving Quantum Computer Problems
Dartmouth College and Sydney University researchers have brought quantum computing a step closer to reality by developing a new method of designing quantum memory. According to these researchers, their method reduces the typically expected number of errors within quantum computing, while not affecting high-speed performances in a detrimental way.
The concept of quantum computing essentially draws on well-established, but none-the-less extremely difficult to control behaviours on subatomic levels. It particularly draws on objects' odd property of simultaneously having negative and positive charges, and the fact that one subatomic particle can affect another without apparently being in physical contact.
Multiplying this over many objects basically affords a far greater amount of computation than currently possible in the digital world of standard zeros and ones. Using these subatomic properties has the potential of substantially increasing power for certain computation types. Implications are thought to be of particular potential use for facial recognition, materials science, code breaking and/ or security.
The extreme difficulty in observing and capturing this type of behaviour is compounded by the fact that such quantum states typically exist for only the tiniest fractions of a second. Keeping quantum information unbroken and relevant to specific computation for prolonged periods is subsequently one of the more discouraging tasks of quantum computing. As it is, the team of researchers believe they are in a position to say that this problem may now be solved.
Using a technique known as dynamic coupling, the team was able to suppress quantum system errors through fluctuations being cancelled out, similar to the way in which a wave can cancel out, or smooth, another, contending wave.
According to the Quantum Control Laboratory's (School of Physics, Sydney University) director, Michael J. Biercuk, waves have the ability to overlap just right in order to create high amplitudes or cancel out undesirable fluctuations. In dynamic coupling, the right kind of interference at just the right time will result in errors to be cancelled.
Existing work within this field was added to by the team through figuring out how a sequence of behaviours can be broken down into smaller segments, enabling preservation of information without distortion of the overall result.
The team was able to show that even if when interrupting after a number of cycles, error probability did not appear to change significantly. This means it should be possible to 'bound' error probability by using repeated dynamic coupling sequences. For system designers, this knowledge of how memory will ultimately perform is naturally of vital importance.
Results of this somewhat significant breakthrough can be found in Nature Communications journal's June 19 issue. Researchers are now required- according to Dr Biercuk - to work on large-scale demonstration of this experimental process on a basis that is repeatable. The next step will then be to integrate this memory system with additional error-correcting algorithms in order to create viable, uniform results.
© Copyright 2021