Error models¶

Error models allow the introduction of probabilistic errors during the execution of the quantum circuit. They are useful for simulating more realistically a real quantum computer.

The error model is specified by adding an error_model statement in the cQasm input, followed by the model name and its parameters.

The only supported error model is currently the depolarizing channel.

Depolarizing channel¶

This model is implemented as described in Quantum Computation and Quantum Information by Nielsen & Chuang. Note that the book uses the density matrix/quantum computation formalism, while currently QX-simulator only uses state vector simulation.

This model is parametrized by a probability of error p. Between each gate of the circuit, an error on a uniformly randomly chosen qubit is applied with probability p. The error is uniformly a X (bit-flip), Y or Z (phase-flip) gate.

For instance, take the following circuit:

version 1.0
qubits 2

error_model depolarizing_channel, 0.0001

h q[0]

measure_all

When simulated 100000 times, it can yield:

Measurement register averaging
     50000/100000 (0.50000000)
     49994/100000 (0.49994000)
     4/100000 (0.00004000)
     2/100000 (0.00002000)