Security Analysis of UAV Swarm Based on Smart Contracts

Abstract

Blockchain technology, characterized by its decentralized nature, traceability, automated operations, and unalterability, is well-suited to address the security requirements of UAV swarms at both theoretical and technical levels. This study delves into identifying potential security loopholes in smart contracts during their deployment, such as integer overflow, timestamp-related issues, reentrancy, transaction order dependencies, and authorization concerns. A novel detection model, integrating a hybrid neural network with an attention mechanism, is introduced to spot these vulnerabilities in smart contracts. Comparative analysis reveals that this model, referred to as ACBSC, outperforms existing popular smart contract vulnerability detection methods. It demonstrates superior accuracy and precision in identifying security flaws. The findings of this research are significant in enhancing the safety of UAV swarm operations and offer valuable insights for the unmanned sector's advancement and development.