We are excited to introduce a new study combining quantum computing techniques with deep learning to strengthen network defense against distributed denial-of-service (DDoS) attacks. DDoS traffic often unfolds as complex temporal patterns that are difficult for classical models to capture efficiently at scale, especially in high-throughput network environments.
The paper introduces a hybrid quantum-classical LSTM architecture, where quantum circuits are integrated into the recurrent learning process to enhance the model’s ability to capture sequential dependencies in network flow data. This hybrid design aims to improve detection sensitivity to evolving attack patterns while exploring how near-term quantum resources can complement classical deep learning for cybersecurity tasks.
The study was authored by A. Kadi, Hajar Moudoud, Lyes Khoukhi, and Zakaria Abou El Houda, and was presented at the 2026 International Conference on Quantum Communications, Networking, and Computing (QCNC) in Kobe, Japan.
Congratulations to the authors for this forward-looking work!