Towards Quantum Resilience: Data-Driven Migration Strategy Design

TL;DR

This paper presents a data-driven decision-support framework using machine learning to guide organizations in transitioning from classical to post-quantum cryptography, addressing quantum threats to security.

Contribution

It introduces a novel, dynamic approach employing decision trees and random forests to recommend mitigation strategies based on system features.

Findings

Key features influencing mitigation strategies identified

Classifier achieves high accuracy in recommending plans

Provides structured roadmap for quantum resilience

Abstract

The advancements in quantum computing are a threat to classical cryptographic systems. The traditional cryptographic methods that utilize factorization-based or discrete-logarithm-based algorithms, such as RSA and ECC, are some of these. This paper thoroughly investigates the vulnerabilities of traditional cryptographic methods against quantum attacks and provides a decision-support framework to help organizations in recommending mitigation plans and determining appropriate transition strategies to post-quantum cryptography. A semi-synthetic dataset, consisting of key features such as key size, network complexity, and sensitivity levels, is crafted, with each configuration labeled according to its recommended mitigation plan. Using decision tree and random forest models, a classifier is trained to recommend appropriate mitigation/transition plans such as continuous monitoring, scheduled transitions, and immediate hybrid implementation. The proposed approach introduces a data-driven and dynamic solution for organizations to assess the scale of the migration, specifying a structured roadmap toward quantum resilience. The results highlight important features that influence strategy decisions and support actionable recommendations for cryptographic modernization based on system context.