Temporal Evolution Tracking Quantum Camouflage Detection Algorithm for Polymorphic Cyber Attacks

his study addresses the challenge of detecting polymorphic and camouflage-enabled cyberat tacks, which rapidly mutate, conceal structural traits, and blend into complex network environ ments. To overcome these , we introduces the Quantum AIO-ChameleonGAN, an advanced gen erative– discriminative framework nesting three complementary components: Quantum Variational Adver- sarial Learning (QVAL), Angle-of-Incidence Optimization (AIO), and Temporal Evolution Tracking (TET). QVAL enhances variational quantum circuits to expand the expressive capacity of feature embeddings, enabling the system to detect subtle spectral and subspace perturbations generated by evolving adversaries. AIO optimizes geometric orientation and incidence sensitivity, allowing the discriminator to capture fine-grained structural distortions common in camouflage driven intrusions. TET introduces temporal regularization to track evolutionary shifts across attack sequences, prevent- ing premature convergence and improving resilience against rapidly shifting threat landscapes.A comprehensive ablation analysis demonstrates that each component contributes uniquely to detec- tion fidelity. Removing QVAL, AIO, and TET results in statistically significant performance degra- dation (paired t-tests and Wilcoxon tests, p < 0.001; Cohen’s d = 0.89–2.10), confirming that quantum expressivity, geometric coherence, and temporal adaptivity independently strengthen ad- versarial robustness. The full QAC-GAN achieves an F1-score of 99.81%, precision of 99.79%, recall of 99.84%, and an anomaly-detection rate of 99.40%, consistently outperforming its classi- cal counterpart (CM-GAN). To further contextualize performance, QAC-GAN was benchmarked against state-of-the-art adversarial attack frameworks, including FGSM and PGD, which represent leading standards for evaluating robustness under perturbations. Whereas FGSM and PGD degrade classifier f idelity to 77–83% across key metrics and frequently induce discriminator collapse, QAC- GAN main tains > 99.7% accuracy, high geometric stability, and strong resilience to gradient-based distortions. This establishes QAC-GAN as a substantially more robust architecture for intrusion and anomaly detection in adversarially manipulated environments.