quant-phNov 12, 2026

No Convergence in Learned Eavesdropping-Detection Dynamics for BB84 Quantum Key Distribution

Authors: Ramon Jose C. Bagunu

Organizations: Department of Physical Sciences and Mathematics, University of the Philippines Manila, Ermita, Manila 1000, Philippines

Abstract

In prior work we showed that a reinforcement-learning eavesdropper can intercept BB84 quantum key distribution while holding the quantum bit error rate below the abort threshold and keeping a usable distillable key. In this paper we ask whether an equally adaptive learning detector can close this stealth window. We cast the interaction as an alternating best-response game between a tabular eavesdropper and a tabular threshold-adapting detector. We measure exploitability, which is the detection rate and distillable key fraction achieved by a fresh attacker best-responding to the final detector. Across three learning rules (Q-Learning, SARSA, Double Q-Learning), three channel-noise levels, and a uniform false-alarm discipline, the dynamics do not settle out to twenty rounds. The exploitability does not approach a fixed point, and its round-to-round swings do not decay. Two controls, one retraining the attacker against a frozen defender and one retraining the defender against a frozen attacker, show that the swings on each side match the variance of the best-response procedure itself. The dynamics fluctuate at the scale of best-response variance rather than converging, with neither player settling to a fixed response. The behavior is noise-dependent, and the stealth window stays open at the low noise where practical QKD operates. The defender contains the attacker on average at intermediate noise without the contest settling, and at high noise the channel rather than the learned policy suppresses the attack. As a candidate mechanism we examine a last-mover information asymmetry, supported by significant anti-persistence at intermediate noise and consistent with best-response noise at low noise. The results suggest that a learned detector cannot be certified from any single round, and that closing the window calls for breaking the information structure rather than out-adapting the attacker.

Figures & tables

Explore similar work

Oct 1, 2026

Learnt Attacks on Quantum Key Distribution under Channel Noise and Device Drift

Quantum key distribution (QKD) links are provisioned from security analyses of stationary channels, whereas the devices that determine the channel drift between recalibrations. Whether an eavesdropper who cannot alter the channel's own noise gains by following that drift has not been quantified. Adaptive eavesdropping is posed here as a constrained Markov decision process in which the attacker selects one circuit per round while the noise level follows an Ornstein--Uhlenbeck process and the abort condition is a budget over each block of rounds. The value of adaptation is bounded by the best fixed circuit and a dynamic-programming upper bound. The actions are learnt attacks. Whereas Decker et al. trained a parametrised circuit on a fixed gate template against a fixed channel, here the gate structure and rotation angles are searched jointly. This yields circuits compact enough to form a discrete action set, extending the construction to noise models lacking a known template, including the amplitude damping channel. On device-independent E91 under bilateral depolarising noise, a reinforcement-learning attacker raises her Holevo information from 0.1350.135 for the best fixed circuit to 0.3480.348 at zero detection, 98%98\% of the upper bound. On BB84 under a drifting bit-flip channel, she exceeds a conservative noise-indexed rule by 0.0240.024 in fidelity, reaching 99%99\% of the upper bound. Under stationary noise, the attacker's gain from basis asymmetry changes sign between an averaged and a per-basis error-rate constraint. The search, started from random gate sequences, recovers the analytical cloners and the collective-attack key rate, and meets the lower bound of the Winick--Lütkenhaus--Coles objective from above.
Aug 4, 2026

Beyond the QBER Threshold: A Temporal QBER Based Machine Learning Framework for Multi Attack Detection in BB84 QKD

Conventional BB84 Quantum Key Distribution (QKD) systems rely on a fixed 11% Quantum Bit Error Rate (QBER) threshold to detect eavesdropping. However, stealthy attacks can remain below this threshold while still compromising channel security. This paper proposes a temporal QBER based machine learning framework for detecting and classifying eavesdropping attacks in BB84 QKD systems. Rather than relying on average session level QBER, the framework extracts 63 physics-informed temporal features capturing burst behavior, temporal instability, basis dependent asymmetry, and QBER loss interactions. Random Forest, XGBoost, and Support Vector Machine with a Radial Basis Function kernel (SVM-RBF) classifiers are evaluated on seven eavesdropping attacks and a normal channel scenario under noisy and lossy conditions. Averaged over ten independent runs, XGBoost achieves the best performance with 88.01% (0.47%) accuracy and a macro F1 score of 0.8803, while SVM-RBF performs comparably, confirming the robustness of the proposed features. Evaluated as a binary attack-versus-normal detector for comparison with conventional monitoring, a fixed 11% QBER threshold achieves only 25.82% accuracy with a False Negative Rate (FNR) of 0.8477, whereas the proposed framework reduces the FNR to 0.0198, substantially improving detection of stealthy attacks that evade threshold-based monitoring. SHapley Additive exPlanations based (SHAP) explainability shows that physics-informed temporal and channel derived features are highly discriminative for identifying eavesdropping strategies. These results demonstrate that temporal QBER driven machine learning provides an accurate, explainable, and practical framework for multi attack security monitoring in BB84 QKD systems.
Oct 3, 2026

Quantum Machine Learning Protection of Military Quantum Key Distribution Against Cryptographically Camouflaged Attacks

Quantum key distribution proves its protocol secure and says nothing about the hardware beneath it, so military and government operators fielding it for command-and-control keys monitor the channel for implementation attacks, and that monitoring has a blind spot. An adversary with a kleptographic foothold in the generator of a public per-block value x=gv(modp)x=g^v \pmod p can hide attacked blocks in honest noise, gating them on a predicate of its discrete logarithm, making detection a discrete logarithm problem that defeats every efficient classical monitor yet yields to a quantum kernel recovering vv through Shor's algorithm. I formalise these cryptographically camouflaged attacks, reduce their hardness to an established learning separation, prove a single-frequency fidelity kernel cannot represent an interval predicate, and test them on Ghillie, a decoy-state BB84 simulator with a positive key rate to 142 km. From 10- to 14-bit groups over two seeds, a classical monitor reads 0.458 to 0.516 on camouflaged attacks while the quantum kernel reads 1.000, and both catch overt attacks above 0.99. Finite-precision recovery under depolarising noise and a hardened predicate lower the quantum result to 0.916 through 0.983 with the classical monitor at chance, and a feasibility probe on IBM Heron processors tracks the exact kernel within 0.034. A defender can therefore discard precisely the compromised key material, although the advantage is asymptotic, awaits fault tolerance, and holds only when the feature map matches the adversary's predicate, since a low-frequency map reads 0.545 on a residue pattern and 0.982 once aligned.