Quantum Reservoir-Based Intrusion Detection for Edge-IoT Systems

The growth of consumer IoT devices has greatly widened the range of cyber threats, creating a demand for intrusion detection systems (IDS) that are accurate, lightweight, and capable of real-time performance. Although edge computing enables localized analysis near data sources, many existing machine learning-based IDS struggle with high computational requirements, costly retraining processes, and limited flexibility in environments with restricted resources. In this work, we propose a novel, quantum-enhanced edge intrusion detection (Q-EDGE-ID), which integrates Quantum Reservoir Computing (QRC) into the edge-IoT landscape.We introduce a system that deals with the natural flow of quantum transition to examine sequential patterns in network and sensor data, all without needing any training in the quantum layer. The simulations show that the proposed system achieves a detection accuracy of 94.1%, an F1-score of 93.5%, and a latency of only 45 ms per inference across various attack scenarios. Additionally, the energy consumption is greatly reduced, requiring just 2.4 mJ for each decision, compared to 6.3–8.7 mJ for CNN- and LSTM-based IDS benchmarks. The system’s robustness is confirmed under noisy conditions, with accuracy staying above 70% even at high noise levels. These findings position Q-EDGE-ID as a lightweight, low-latency, and energy-efficient intrusion detection solution suitable for next-generation IoT deployments.