I want to replicate the depolarizing noise channel

enter image description here

for a 4 qubit circuit system, where p is the probability for an error. enter image description here I tried doing this: enter image description here

But I get the error WARNING: all-qubit error already exists for instruction "measure", composing with additional error.

I have looked at Qiskits documentation on the depolarizing_error function but it is not clear, or I don't understand the maths.

Could anyone please shed more light on this?

Edit: I might have found a solution, but I'm not sure if my logic is right.

For my 4 qubit system, I want to apply the depolarizing channel noise to the measurement of the qubits $q_{0}$ and $q_{2}$ in the z-direction:

enter image description here

Edit: I realized that the single-qubit depolarizing channel above acts separately on the two qubits: enter image description here


1 Answer 1


The aim is to use Qiskits built in depolarizing quantum error channel function $\texttt{depolarizing}$$\_$$\texttt{error(param, num_qubits, standard_gates=True)}$ to create the desired depolarizing channel. It takes the depolarization parameter $\lambda$ and number of qubits $n$ to create the depolarizing channel $$ \begin{equation} \label{eqn:depo-chan-qiskit} \varepsilon(\rho) = (1-\lambda)\rho + \lambda Tr\left[\rho\right]\frac{I}{2^{n}} \text{.} \end{equation}$$

It can then be manipulated to create the single qubit depolarizing channel $$ \begin{equation} \label{eqn:depo-chan-1qubit} \varepsilon(\rho_{single-qubit}) = (1-P)\rho + \frac{P}{3}\left(X\rho X + Y\rho Y + Z\rho Z\right) \text{.} \end{equation}$$ This is shown by the following derivation. For a single qubit $n=1$ $$ \frac{I}{2^{1}} = \frac{1}{4} \left(I\rho I + X\rho X + Y\rho Y + Z\rho Z\right),$$ and $Tr[\rho] = 1.$

If we then chose $$\lambda=\frac{4^{n}P}{4^{n}-1},$$ where $P$ is the probability. At $n=1$ the depolarization paramater $$\lambda = \frac{4P}{3}$$ and by substituting these results into $\varepsilon(\rho)$ we get that

$$ \begin{align} \varepsilon(\rho) &= (1-\frac{4P}{3})\rho + \frac{4P}{3} \left(I\rho I + X\rho X + Y\rho Y + Z\rho Z\right) \frac{1}{4} \\ &= \left( 1 - \frac{4P}{3} + \frac{P}{3} \right)\rho + \frac{P}{3} \left(X\rho X + Y\rho Y + Z\rho Z\right) \\ &= (1-P)\rho + \frac{P}{3}\left(X\rho X + Y\rho Y + Z\rho Z\right) \text{.} \end{align} $$ enter image description here

Here is a link to my notebook explaining how it works https://github.com/MIGUEL-LO/Qiskit_depolarazation_channel/blob/master/depo_channel_using_depolarizing_error.ipynb


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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