Quantum Permutation Pad with Qiskit Runtime

Abstract

We demonstrate an efficient implementation of the Kuang and Barbeau’s Quantum Permutation pad (QPP) symmetric cryptographic algorithm with Qiskit Runtime, a new architecture offered by IBM Quantum that streamlines quantum computations. We have implemented a Python class QPP and template Jupyter notebooks with Qiskit code for encrypting and decrypting with n-qubit QPP any text file in UTF-16 format or any image file in .png format. We offer the option of running either a quantum circuit with n qubits, or an alternate one with 2ⁿ qubits which only uses swap gates and has a circuit depth of O(n). It is inherently extremely fast and could be run efficiently on currently available noisy quantum computers. Our implementation leverages the new Qiskit Sampler primitive in localized mode which dramatically improves performance. We offer a highly efficient classical implementation which performs permutation gate matrix multiplication with information state vectors. We illustrate the use with two agents Alice and Bob who exchange a text file and an image file using 2-qubit QPP and 4-qubit QPP.