Ensuring Secure Transmission in VANET: Optimal Clustering and Improved LSTM‐Based Intrusion Detection

ABSTRACT VANETs are essential for future intelligent transportation systems, enabling communication among vehicles and infrastructure. However, their decentralized nature presents significant security challenges. This paper proposes a holistic framework to address these issues, comprising three components. First, the Self‐Improved Kookaburra Optimization Algorithm (SIKOA) optimizes cluster formation by considering trust, mobility patterns, and link lifetime, enhancing network efficiency and resilience. Second, an Improved Long Short‐Term Memory (LSTM) model is introduced for intrusion detection, incorporating link quality and trust metrics to minimize false positives. Finally, an Improved Blowfish Algorithm is applied for data encryption, ensuring confidentiality and integrity. Experimental results in simulated VANET environments show that the proposed framework outperforms existing solutions in terms of detection accuracy, computational efficiency, and attack resilience.