Reputation-Enhanced Practical Byzantine Fault Tolerance Algorithm for Node Capture Attacks on UAV Networks

Abstract The rapid growth of the Unmanned Aerial Vehicles (UAVs) market has led to increased reliance on UAV technologies across various application domains such as aerial Internet of Things (IoT). However, the vulnerability of UAV infrastructures to security threats and attacks poses significant risks. The discovery of robust security measures is particularly crucial for UAVs operating in hostile environments. This paper proposes a Reputation-enhanced Practical Byzantine Fault Tolerance Algorithm (RePA) to address node capture attacks on UAV networks in adversarial settings. RePA integrates a reputation system into the consensus process, assigning master and candidate nodes based on their reputation values, thereby reducing the likelihood of malicious nodes becoming decisive. RePA also introduces a replica node that serves as a backup to master candidacy, enhancing resilience against node capture attacks. This study advances UAV network security by demonstrating the effectiveness of combining reputation scores with PBFT as a defense mechanism against node capture attacks in hostile environments. An experimental study validates the efficacy, reliability, and performance of RePA.