Related Experiment Video
Updated: Aug 6, 2026
05:30
Large Scale Energy Efficient Sensor Network Routing Using a Quantum Processor Unit
Published on: September 8, 2023
A quantum enhanced neuro symbolic intrusion detection system for software defined networking
D Lynda1, G Logeswari2, K Tamilarasi1
1School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, Tamil Nadu, 600 127, India.
Scientific Reports
|July 16, 2026
Summary
NeuroTwin QIDS, a quantum-enabled intrusion detection system for Software Defined Networking (SDN), enhances network security by accurately identifying and classifying anomalies. This novel approach significantly improves upon existing intrusion detection systems (IDSs).
Area of Science:
- Cybersecurity
- Quantum Computing
- Network Security
Background:
- Software Defined Networking (SDN) offers programmability and centralized control but expands the attack surface, challenging traditional Intrusion Detection Systems (IDSs).
- Novel attack vectors in SDN environments necessitate advanced detection mechanisms beyond classical IDSs.
- Existing IDSs struggle to effectively address the unique vulnerabilities introduced by SDN architectures.
Purpose of the Study:
- To propose NeuroTwin QIDS, a novel quantum-enabled neuro-IDS specifically designed for SDN environments.
- To enhance anomaly detection capabilities in SDN by leveraging quantum computing and advanced neural network architectures.
- To ensure the statistical relevancy and defensibility of features used for anomaly representation in SDN.
Main Methods:
- Implemented a Neuro-Symbolic Feature Pruning (NSFP) approach using Graph Attention Networks (GAT) and policy-aware symbolic pruning.
- Mapped statistical features to Quantum Reservoir (QRE) space for enriched temporal classification.
- Utilized a two-step classification process: Quantum Reservoir-Transformer Hybrid (QRT-Net) for real-time detection and Capsule Network-BiLSTM (Cap-BiLSTM) for behavior-based classification.
- Integrated neuro-symbolic fusion and a digital twin model for prediction validation and simulation of protective actions.
Main Results:
- NeuroTwin QIDS demonstrated outstanding detection performance, surpassing state-of-the-art IDS systems on benchmark datasets.
- Achieved 99.12% accuracy, 98.95% precision, 98.87% recall, and 98.91% F1-score on the CSE-CIC-IDS2018 dataset.
- Attained 98.67% accuracy, 98.42% precision, 98.15% recall, and 98.28% F1-score on the InSDN dataset.
Conclusions:
- NeuroTwin QIDS effectively addresses the challenges of intrusion detection in SDN environments.
- The proposed quantum-enabled neuro-IDS offers a significant advancement in network security.
- The system's ability to validate predictions against SDN policies and simulate protective actions enhances network self-healing capabilities.