2
$\begingroup$

I'm John and I have a question which I have been thinking about. I'm studying quantum information, especially, quantum error-correcting codes. When I learned some types of quantum codes (e.g. 5 qubits code or quantum Hamming code), I thought the following condition is also a necessary condition for quantum codes, but I have no idea to prove it. Could anyone prove it although it might be false?

Let $\mathbb{P}=\{P_1,...,P_n\}$ be a projection measurement, $\rho$ an $m$ qubits codeword and $U_i$ a unitary operator which represents an error ($i=1,...,m$:position of error). Then, the condition is $$\forall k_1,k_2\in \{1,...,n\}, \forall i\in \{1,...,m\},\mathrm{Tr}(P_{k_1}U_i\rho U_i^\dagger)\neq 0 \land \mathrm{Tr}(P_{k_2}U_i\rho U_i^\dagger)\neq 0\\ \Rightarrow \frac{P_{k_1}U_i\rho U_i^\dagger P_{k_1}}{\mathrm{Tr}(P_{k_1}U_i\rho U_i^\dagger)}=\frac{P_{k_2}U_i\rho U_i^\dagger P_{k_2}}{\mathrm{Tr}(P_{k_2}U_i\rho U_i^\dagger)}$$

$\endgroup$
1
$\begingroup$

I don't believe that this is true (at least without further qualifiers). If I'm correctly understanding what you're trying to convey, I think we can essentially simplify this statement to asking whether there can exists a state $|\psi\rangle$ and two projectors $P_1$ and $P_2$ such that $$P_1|\psi\rangle\neq P_2|\psi\rangle$$ (ignoring normalisation for simplicity). If so, your statement is false.

Now, let $|\phi\rangle$ be a state that is orthogonal to $|\psi\rangle$. Then we can choose $$ P_1=\frac{1}{2}(|\psi\rangle+|\phi\rangle)(\langle\psi|+\langle\phi|),\qquad P_2=\frac{1}{2}(|\psi\rangle-|\phi\rangle)(\langle\psi|-\langle\phi|) $$ Now if you want to start imposing specific properties of syndrome measurements for error correcting codes, that might be a different story... (even then, you might want to distinguish between non-degenerate codes and degenerate codes. Without much thought, I guess your statement is true for syndrome measurements on a non-degenerate code).

| improve this answer | |
$\endgroup$
  • $\begingroup$ Thanks for your answer. I think you are right and I'm wrong. $\endgroup$ – John Nov 15 '19 at 10:24

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.