Novel Quantum Ensemble Machine Learning Models for IoT Intrusion Detection

The rapid growth of Internet of Things (IoT) devices has created significant cybersecurity risks. Current quantum computer-based methods for detecting digital intrusions often rely on uniform data processing techniques that fail to fully utilize the power of quantum mechanics. To address this, we introduce two new quantum ensemble methods: Quantum Voting Ensemble (QVE) and Quantum Weighted Ensemble (QWE). Unlike traditional methods that vary the data or the model structure to improve accuracy, our approach creates diversity by projecting the same data into multiple, geometrically different quantum environments. When tested on a standard IoT security dataset, our QVE method achieved 99.5% accuracy, nearly matching the performance of high-end traditional computers while significantly outperforming existing quantum models. These results show that focusing on how data is encoded into quantum states can create a more robust defense for critical security applications.