基于人工智能的物联网边缘节点间数据流零信任管理系统  

Zero-Trust Management System for Data Flow Among Internet of Things Edge Nodes Based on Artificial Intelligence

作　　者：朱丽 ZHU Li(Wuchang Vocational College,Wuhan 430202,China)

机构地区：[1]武昌职业学院,武汉430000

出　　处：《智能物联技术》2025年第1期80-83,共4页Technology of Io T& AI

摘　　要：针对边缘计算环境下的数据流安全管理,提出基于人工智能的零信任管理机制,通过多维信任评估体系和深度神经网络实现动态评估,结合分级授权与异常检测,实现细粒度安全管控。NS-3仿真验证表明,所提机制在50节点场景下平均响应延迟为42.1ms,对分布式拒绝服务攻击(Distributed Denial of Serviceattack,DDoS)的检测率达97.8%,误报率为2.3%,在信任阈值0.7时实现性能与安全性的最优平衡,具备有效的安全防护能力。Aiming at data flow security management in edge computing environment,a zero-trust management mechanism based on artificial intelligence is proposed.Multi-dimensional trust evaluation system and deep neural network are used to achieve dynamic evaluation,and hierarchical authorization and anomaly detection are combined to achieve fine-grained security control.NS-3 simulation results show that the proposed mechanism has an average response delay of 42.1 ms in the scenario of 50 nodes,and the detection rate of Distributed Denial of Service attack(DDoS)is 97.8%and the false positive rate is 2.3%.When the trust threshold is 0.7,the optimal balance between performance and security is achieved,and effective security protection is provided.

关 键 词：物联网 边缘计算 零信任 深度学习 访问控制 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]