QML-based Intrusion Detection System for 3GPP 6G Traffic

5G and 6G core networks are built on cloud-native, microservices-based architectures, exposing a wide and constantly growing attack surface across network functions (AMF, SMF, UDM, UPF). Classical AI/ML-based intrusion detection is reaching its scalability and expressiveness limits as traffic volume, attack sophistication, and feature-space complexity increase. This thesis explores Quantum Machine Learning as a structurally different computational paradigm to overcome these limits, through three coupled challenges:

1. Architecture: designing a QML-IDS that integrates natively into the 5G/6G Service Based Architecture, compliant with 3GPP Technical Specification, including TS 23.501 (system architecture), TS 33.501 (security architecture), and the emerging AI/ML framework defined in TS 23.288.
2. Proof of Concept: implementing the QML-IDS on the NGN team's 5G core testbed, integrating quantum simulation frameworks (PennyLane, Qiskit, Cirq) under real-time constraints.
3. Algorithm design: developing and evaluating QML-based detection algorithms across CIoT, eMBB, URLLC, V2X, and prospective 6G traffic classes, benchmarked against classical baselines.

Company Description

UVSQ Université de Versailles Saint-Quentin-en-Yvelines is a founding member of Université Paris-Saclay, a top-ranked French institution recognized internationally, including a leading position in atmospheric sciences. Located in the Yvelines department, UVSQ is a major hub for higher education, research, and technology, offering over 200 degree programs across four broad fields: arts and humanities, social sciences, law and economics, and science, technology, and health. Its five campuses host around 20,000 learners, supported by a large community of faculty and doctoral researchers. With 38 research structures and strong emphasis on innovative, interdisciplinary projects, UVSQ actively collaborates with industry through major national investment programs, scientific and industrial chairs, and multiple technological platforms. The university maintains extensive international partnerships in about 50 countries and is involved in several globally visible competitiveness clusters.

About the Laboratory

This PhD will be conducted within the DAVID Laboratory (Données et Algorithmes pour une Ville Intelligente et Durable) at UVSQ, Université Paris-Saclay, in the Next Generation Network (NGN) research team. The NGN team specializes in the security and architecture of cloud-native cellular networks (4G/5G/6G) and has pioneered the integration of Intrusion Detection Systems directly as Network Functions inside the 5G core, notably through the 5G-IoT-IDS system and its extension to IP/Non-IP CIoT traffic, both validated on an operational 5G core testbed (Open5GS, free5GC) built and maintained by the team. This PhD builds directly on that research baseline to introduce Quantum Machine Learning into the next generation of in-core intrusion detection

We are looking for a candidate who is genuinely motivated to pursue a PhD and a research career.

The candidate should hold a Master’s degree in Computer Science, Telecommunications, Computer Engineering, or a closely related field.

Good knowledge of the following areas is expected:

• Computer networks and telecommunications
• 5G/6G networks
• Machine Learning / Artificial Intelligence
• Programming and software development
• Applied mathematics, particularly linear algebra, probability and optimization

A strong academic record throughout the university curriculum will be highly appreciated.

The following are considered additional assets, but are not mandatory requirements:

• A Master’s degree completed in France;
• A Master’s internship carried out in France;
• Previous experience with 5G/6G networks;
• Experience in Machine Learning applied to networking or cybersecurity;
• Knowledge of Quantum Machine Learning;
• Familiarity with PennyLane, Qiskit or Cirq.