FAPL-DM-BC: A Secure and Scalable FL Framework with Adaptive Privacy and Dynamic Masking, Blockchain, and XAI for the IoVs

Adaptive privacy protection responds to real-time data sensitivity: The framework automatically adjusts privacy settings based on how sensitive vehicle data is at any given moment, rather than using fixed privacy levels. This allows the system to maintain stronger protection when needed while optimizing performance when data is less sensitive.

Blockchain provides transparent accountability for vehicle data: By using blockchain-based record-keeping and decentralized validation, the system creates an auditable trail of how vehicle data is processed and shared. This transparency helps ensure that no single entity can secretly misuse data from connected vehicles.

Explainable AI makes autonomous vehicle decisions understandable: The framework incorporates explainable AI tools that can show why the system made specific decisions (such as traffic predictions or safety alerts), rather than operating as a "black box." This transparency is critical for building public trust in autonomous vehicles and smart city systems.

Multiple security layers protect against different attack types: The system combines secure data aggregation, encrypted multi-party computation, and secure logging to defend against various threats. This layered approach is important because IoV systems face diverse cybersecurity risks from multiple potential attackers.

Decentralized learning reduces privacy risks compared to centralized systems: By processing vehicle data locally and only sharing aggregated insights (rather than raw data), the framework minimizes the amount of sensitive information that must be transmitted or stored centrally, reducing the impact of potential data breaches.