Research

This ongoing research focuses on building a high-performance intrusion detection framework for IoT environments by first replicating and validating a deep learning–based Hybrid LSTM-CNN model using a fair, paper-faithful training pipeline across CNN, RNN, LSTM, BiLSTM, GRU, and Hybrid LSTM-CNN architectures. After identifying the best-performing classical model, the study extends the work by integrating Quantum Computing concepts into the strongest architecture to explore whether quantum-enhanced learning can further improve threat detection, robustness, and efficiency. The overall goal is to create a reliable, scalable, and interpretable IoT security system capable of detecting cyberattacks such as DDoS, reconnaissance, botnet activity, and data exfiltration with high accuracy while maintaining fairness in model comparison and practical feasibility under limited hardware resources.