The Shor code can be used to protect against an arbitrary one-qubit error. If the time interval over which quantum noise acts is short enough, can it be assumed only one-qubit errors occur, so one does not have to deal with two-qubit errors, etc.


That might depend on your noise model. A typical noise model is independent errors on each qubit, occurring with probability $p$. In that case, as soon as there's a non-zero chance of having a single-qubit error, there's a non-zero chance of have a two-qubit error. But it could be a negligibly small probability.

Part of the point of an error correcting code is only really to reduce the probability of error. A single qubit has a probability of error $p$ while a qubit encoded in the Shor code has a probability of error of $$ 1-(1-p)^9-9p(1-p)^8\approx 36p^2-168p^3 $$ So provided $36p^2<p$, you're gaining some benefit. Encode it again, and you're doing even better. Keep on encoding, and you can get arbitrarily accurate.

  • $\begingroup$ If $p$ is small you might argue that you can neglect the $\mathcal{O}(p^2)$ terms. $\endgroup$
    – M. Stern
    Sep 9 at 6:55
  • $\begingroup$ @M.Stern Agreed, hence my comment "could be a negligibly small probability". But without any external criterion, I don't know if I can neglect those terms. $\endgroup$
    – DaftWullie
    Sep 9 at 8:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.