Hybrid Quantum-CNN Framework for Secure, Robust, and Efficient IoT Localization under Adversarial Signal Attacks

Accurate and secure localization is critical for mission-critical IoT applications, yet networks remain vulnerable to attacks such as Neutralization-Inspired Fake Signal (NIFS) attacks. We address accurate positioning under adversarial conditions while respecting latency and energy constraints. We formulate a secure localization problem optimizing accuracy, cryptographic overhead (Ed25519), and robustness against malicious anchors, and propose a Hybrid Quantum-Convolutional Neural Network (HQC-NN) with K-Means clustering for region-specific specialization. The framework integrates cryptographic verification and monitors performance, energy, and security metrics. Simulations show that HQC-NN and HQC-NN-KMeans outperform CNN and CNN-KMeans baselines in accuracy, efficiency, latency, and attack detection, demonstrating their potential for resilient IoT localization.