Quantum-Enhanced Ddos Detection in 5G Cloud Networks using Bottleneck Attention Mechanism

Abstract

The rapid expansion of 5G networks and cloud computing has heightened the risk of “Distributed Denial of Service (DDoS)201D attacks, which can severely compromise service availability and network performance. Traditional machine-learning techniques have shown limitations in accurately detecting these evolving attacks under high-traffic conditions, especially in 5G environments. This research proposes an advanced DDoS detection framework utilizing a “Quantum Convolutional Neural Network (QCNN)” combined with a Bottleneck Attention Mechanism. The QCNN extracts spatial and temporal features from network traffic, while the Bottleneck Attention Mechanism prioritizes critical patterns, improving accuracy and computational efficiency. The framework is evaluated using the 5G Non-IP Data Delivery (NIDD) and CIC-DDoS2019 datasets to ensure robustness across diverse attack scenarios. Data preprocessing techniques, including feature engineering and normalization, are employed to prepare the datasets for optimal model performance. The proposed model performs better than conventional profound learning models, such as “CNN, LSTM, and Bi-LSTM” in terms of accuracy, precision, recall, and F1 score. This study demonstrates that integrating quantum-inspired learning with attention mechanisms significantly enhances DDoS detection capabilities, making it an effective solution for safeguarding 5G-enabled cloud environments. The findings emphasize the importance of advanced architectures for real-time, high-precision DDoS detection, essential to ensuring future cloud infrastructures' security and reliability.