Vehiguard Transfer Learning-Driven Quantum PSO-LSTM for Automotive Cyber Security

Quantum-Inspired Particle Swarm Optimization (QPSO) has emerged as a powerful optimization technique that enhances hyper parameter tuning and convergence efficiency in machine learning models. In the context of vehicular network security, integrating QPSO with Long Short-Term Memory (LSTM) networks and Transfer Learning offers a robust framework for detecting cyber threats in dynamic vehicular environments. Traditional anomaly detection mechanisms often struggle with real-time adaptability, leading to reduced detection accuracy and delayed response to evolving attack vectors. The proposed VehiGuard framework leverages QPSO to optimize hyper parameters efficiently, improving LSTM’s ability to detect anomalies in vehicular data streams. Transfer learning further enhances model adaptability by utilizing pre-trained models from related domains, significantly reducing the need for extensive labelled datasets while maintaining high detection accuracy. This hybrid approach ensures a scalable and computationally efficient solution for vehicular cyber security. Empirical evaluations conducted on widely used vehicular datasets demonstrate that VehiGuard outperforms conventional models in terms of detection accuracy, computational efficiency, and robustness against diverse attack vectors. The results highlight the potential of integrating quantum-inspired optimization with deep learning and knowledge transfer to secure autonomous vehicles and IoT-enabled transportation systems. By addressing critical challenges in vehicular network security, VehiGuard contributes to the development of more resilient and adaptive intelligent transportation systems.