qcostas

Quantum algorithm for generating Costas arrays with Grover’s search (Qiskit).
It builds a Boolean oracle for the Costas constraints, runs Grover, decodes a permutation, verifies the Costas property, and plots the result. Best suited for small n (the oracle grows quickly).

Install

pip install -r requirements.txt

Quick start

from qcostas import generate_costas, plot_costas
import matplotlib.pyplot as plt

n = 3
res = generate_costas(n)

print("Found permutation:", res.permutation, "Costas?", res.is_costas)
plot_costas(res.permutation)
plt.show()

You can also open and run example.ipynb.

What you get

generate_costas(n) returns a CostasResult dataclass:

• permutation: list[int] - 1‑indexed columns (one per row)
• is_costas: bool - sanity check of the Costas property
• N, M, r: int - search space size, target count estimate, Grover iterations
• grover, problem, result - Qiskit objects if you want to inspect the circuit/results

To visualize, call plot_costas(permutation).

Notes & limits

• This is a research/educational demo. Grover offers quadratic speedup, but the oracle (Costas constraints) makes circuits large; use small n.
• By default it uses Qiskit’s StatevectorSampler (simulator). You can swap in another sampler if you want hardware/backend runs.

License

MIT — see LICENSE.