Qubit-efficient quantum intrusion detection using dual-parameter encoding and multi-metric calibration

Area of Science:

• Quantum Computing
• Cybersecurity
• Network Intrusion Detection

Background:

• Modern networks generate high-dimensional data, challenging traditional intrusion detection systems.
• Quantum computing offers potential for enhanced detection but faces qubit limitations.
• Effective cybercrime defence requires efficient feature extraction and anomaly detection.

Purpose of the Study:

• To present a quantum-enhanced intrusion detection framework addressing qubit limitations.
• To introduce Dual-Parameter encoding for efficient feature representation.
• To develop a multi-metric anomaly scoring system with calibrated false-positive control.

Main Methods:

• Employed Dual-Parameter encoding using qubit polar angle (θ) and azimuthal phase (ϕ) to encode two normalized features.
• Developed a composite anomaly score unifying quantum fidelity, Kullback-Leibler divergence, and von Neumann entropy change.
• Utilized quantile calibration and the Dvoretzky-Kiefer-Wolfowitz inequality for false-positive rate control.

Main Results:

• Dual-Parameter encoding reduced data-qubit requirements from [Formula: see text] to [Formula: see text] for a feature set of size [Formula: see text].
• Evaluated the Dual-Parameter Quantum Intrusion Detection System (DP-QIDS) on UNSW-NB15 and CIC-IDS2017 datasets.
• Demonstrated that the fusion of fidelity, KL divergence, and entropy ensures reliable detection of various attack types.

Conclusions:

• DP-QIDS advances quantum intrusion detection towards practical, resource-aware deployment.
• The framework provides auditable and resilient security operations for modern network infrastructure.
• The study connects quantum simulation with real-world security applications.