Noise Models - Noro-Official/OQS GitHub Wiki

Noise Models in OpenQStack

Quantum information is fragile—errors arise from interaction with the environment, imperfect gates, measurement collapse, and more.

OpenQStack provides a modular framework to simulate these effects with realistic, customizable quantum noise channels.

Built-In Error Channels

Bit-Flip Channel

Simulates a random Pauli-X error on a single qubit.

from openqstack.qec import BitFlipChannel

channel = BitFlipChannel(p=0.1, n_qubits=3, target_qubit=1)
noisy_state = channel(state)

Phase-Flip Channel

Simulates a Z error with probability p.

from openqstack.qec import PhaseFlipChannel

Depolarizing Channel

Applies X, Y, or Z randomly with equal probability (total probability p):

from openqstack.qec import DepolarizingChannel

Channel API

All noise channels are subclasses of ErrorChannel and support:

  • apply(state, mode="stochastic") — default: single-shot Kraus sampling
  • apply(state, mode="average") — applies the full density matrix evolution

You can also pass channels as callable functions:

state = channel(state, mode="average")

Custom Channels

You can define your own noise model by subclassing ErrorChannel:

class MyCustomNoise(ErrorChannel):
    def kraus_operators(self):
        # Return list of (K_i, p_i) pairs
        return [...]

Examples include:

  • Amplitude damping
  • Correlated noise
  • Crosstalk
  • Leakage or dephasing

Testing Noise Effects

Try modifying openq_blink.py to inject different noise models and compare logical error rates after recovery.

You can also visualize impact with:

from openqstack.visualize import plot_bloch
plot_bloch(decoded)

Coming Soon

  • Multi-qubit entangled noise
  • Circuit-level error insertion
  • Benchmarking with QECBench
  • Visualization of noise map evolution

Related Pages