Original URL: http://www.theregister.co.uk/2012/05/21/noisy_quantum_computer/

## Noise can improve quantum computing, says ANU scientist

It’s quantum computing: of *course* there’s a paradox

Posted in Science, 21st May 2012 00:47 GMT

Here’s a nice paradox: since noise gets in the way of quantum computing, cure it by adding more noise.

That startling proposal is the work of a team led by the Australian National University’s Dr André Carvalho, along with collaborators from Brazil and Spain.

Noise is normally treated as the enemy in quantum-level experiments, because it destroys the useful characteristics of qubits (quantum computing elements). As Dr Carvalho explained to *The Register*, “in the quantum world, these operations work because the qubits can be in a state of superposition” (that is, multiple possible states existing at the same time – as did the famous cat belonging to Herr Schrödinger').

“Those coherence properties disappear with time, and that means the main resource, entanglement, disappears with noise.”

The particular kind of noise Carvalho’s work deals with is the spontaneous emission of photons. This means, simply, that if a qubit starts in a “1” state, it will eventually emit a photon and fall to a “0” state.

Without a solution to the noise problem, Dr Carvalho says, computation becomes impossible. “Because we have no control on the outcomes of the measurement – they are totally random – if we just passively wait it would take an infinite amount of time to extract even a very simple computation.”

Dr Carvalho’s solution is two-fold: first, to add noise (as photons, using a laser) back into a qubit *before* the state-decay takes place; second, to perform measurement on the system in just the right way.

If, after adding the extra noise, the system were simply left alone, decoherence will happen more quickly, he said. “But if we measure the photons that are coming out, and measure them in the right way, then at the end of the measurement, we have a quantum gate.”

In other words, under the right circumstances, the act of measurement is what creates the quantum gate that can perform computation.

Unfortunately, measurement, like noise, is an enemy of entanglement and superposition. Just as in Schroedinger’s thought experiment, if you “look in the box”, the quantum system will resolve itself into a classical state.

“When you measure a quantum system, you destroy the system,” he said. “If you measure the state of a photon, you will know that it’s in a particular state.”

However, because of the added noise, he said, there are two kinds of photons available for measurement – those that are created by spontaneous emission, and those added in the excitation process.

“We have two kinds of photons, the spontaneous and the ‘noisy channel’,” he said – and since the detector “can’t tell whether the photon is coming from one process or the other, that creates a superposition.”

If you want a more metaphorical description of the process, the ANU puts it like the “million monkeys typing Shakespeare” problem: it becomes much more likely if you capture keystrokes that appear in the correct order.

Dr Carvalho's work has been published in *Physical Review Letters*. ®