Quantum-Powered Cyber Attack Detection and Prevention System

Author

T. Debanga Saikia, P. Abhigna, P. Charan Teja, L. Harika, P. Samuel

Abstract

 The increasing complexity of cyberattacks creates a need for intelligent systems capable of accurate detection, explainable analysis, and adaptive response. This paper presents a Quantum-Powered Explainable Cyber Attack Detection and Prevention System, Q-XCADPS, an enterprise Security Operations Center that integrates classical machine learning, quantum machine learning, explainable artificial intelligence, and adaptive response mechanisms. The system detects Distributed Denial-of-Service attacks, SQL Injection, malware, brute-force attacks, and phishing URLs. Random Forest and XGBoost are evaluated along with Quantum Support Vector Classifier and Variational Quantum Classifier using a four-qubit architecture with PCA-based dimensionality reduction. Explainability is provided using SHAP-based feature analysis, while the adaptive response engine assigns threat scores and severity levels. Experimental results show that XGBoost achieves the highest performance with 90.50 percent accuracy and a weighted F1-score of 0.8651. The quantum models demonstrate the feasibility of quantum machine learning for cybersecurity detection.

Keywords

Cybersecurity, Quantum Machine Learning, Explainable AI, Cyber Attack Detection, Q-XCADPS, QSVC, VQC, XGBoost, SHAP.

Full Text:

Download Paper PDF

References

[1] Danasabe, A., Afaq, M., El-Maleh, A. H., Khan, A., El-Ferik, S., & Kaleem, Z. (2026). QATNet: A lightweight quantum-classical tabular network for low-latency intrusion detection. IEEE Transactions on Quantum Engineering, 7, 3102015.

[2] Hdaib, M., Rajasegarar, S., & Pan, L. (2024). Quantum deep learning-based anomaly detection for enhanced network security. Quantum Machine Intelligence, 6(1), 26.

[3] Kukliansky, A., Orescanin, M., Bollmann, C., & Huffmire, T. (2024). Network anomaly detection using quantum neural networks on noisy quantum computers. IEEE Transactions on Quantum Engineering, 5, 3100611.

[4] Kalinin, M., & Krundyshev, V. (2022). Security intrusion detection using quantum machine learning techniques. Journal of Computer Virology and Hacking Techniques, 19(1), 125–136.

[5] Cirillo, F., Esposito, C., & Seo, J. T. (2026). Benchmarking quantum machine learning methods for intrusion detection on noisy quantum computers. Quantum Machine Intelligence, 8(1), 27.

[6] Adenusi, D. A., Oladimeji, O. O., Oyekola, T. A., & Olagunju, K. S. (2025). Data-driven network intrusion detection using optimized machine learning algorithms.

[7] Mohammad, A., & Mohammad, H. (2026). DEHOA-optimized deep intrusion detection with quantum neural network extensions for IoT network security.

[8] Vinotha, K., & Eswaran, P. (2026). Quantum-inspired hyperparameter optimization for enhanced deep learning-based intrusion detection in wireless sensor networks.

[9] Amara, M., Amara, M., Mnasri, S., & Val, T. (2025). QCNN-ID: A quantum-classical hybrid model for IoT intrusion detection.

[10] Kadry, H., Farouk, A., Zanaty, E. A., & Reyad, O. (2023). Intrusion detection model using optimized quantum neural network and elliptical curve cryptography for data security.

[11] Ahmed, S., Shihab, I. F., & Khokhar, A. (2025). Quantum-driven zero trust architecture with dynamic anomaly detection in 7G technology: A neural network approach.

[12] Mahesha, A., Aa, N., Shivvanii, S., Ca, V., & Manimaran, S. (2026). An explainable machine learning and deep learning framework for malicious URL detection.

[13] Dirchaoui, M., Abarda, A., & El Ouardighi, A. (2026). Detecting fraudulent URLs on Arabic social platforms using an explainable machine learning approach. Procedia Computer Science, 280, 1306–1311.

[14] Doshi, T., Patel, V., Shah, N., Swain, D., Swain, D., & Acharya, B. (2025). PhishHunter-XLD: An ensemble approach integrating machine learning and deep learning for phishing URL classification.

[15] Prasad, Y. B., Dondeti, V., & Dondeti, R. (2026). Phishing URL detection in real-time: Machine learning and deep learning approaches on a custom threat intelligence dataset. Procedia Computer Science, 283, 4859–4879.