摘要
云原生列控应用在开放计算环境中面临逻辑劫持、时序攻击和拜占庭失效等安全挑战,为此,提出一种逻辑-时序双维度可信验证架构(Logic-Temporal Dual-dimension Verification,LTDV)。该架构通过Paillier同态加密构建逻辑证据链以验证程序控制流的完整性,利用时序哈希链绑定时间戳以确保行为的实时性,并基于结果驱动的过滤式多数表决机制实现轻量级的拜占庭容错。在云化列控仿真平台的临时限速场景实验中,结果表明该架构能够有效检测多种典型攻击场景,平均检测率达97.8%,在正常工况测试中,系统误报率低于0.1%。与此同时,系统时延稳定在百毫秒级别,确保开销可控。总体而言,该架构在毫秒级响应时间内显著提升云端列控计算的可信性,具有良好的工程应用前景。
Cloud-native train control applications face security challenges such as logic hijacking,timing attacks,and Byzantine failures in open computing environments.To address these issues,this paper proposes a Logic-Temporal Dual-Dimension Verifi cation(LTDV)architecture.The proposed architecture leverages Paillier homomorphic encryption to construct a logic evidence chain for verifying the integrity of program control fl ow,employs a temporal hash chain to bind timestamps for ensuring real-time behavior,and implements a result-driven filtering-based majority voting mechanism to achieve lightweight Byzantine fault tolerance.Experiments conducted on a cloud-based train control simulation platform for temporary speed restriction scenarios demonstrate that the proposed architecture eff ectively detects various types of typical attack scenarios,achieving an average detection rate of 97.8%and a false positive rate of less than 0.1%under normal operating conditions.Meanwhile,the system latency remains stable at the hundreds of milliseconds level,ensuring controlled overhead.Overall,the proposed architecture signifi cantly enhances the trustworthiness of cloud-based train control computing within the constraints of millisecond response times,offering promising prospects for practical engineering deployment.
作者
芦泽和
朱力
梁雅楠
Lu Zehe;Zhu Li;Liang Yanan(Beijing Jiaotong University,Beijing 100044,China)
出处
《铁路通信信号工程技术》
2026年第2期10-16,共7页
Railway Signalling & Communication Engineering
基金
国家重点研发计划项目(2024YFB3108600)
北京市自然科学基金重点项目(L251011)。
关键词
云端列控
临时限速
逻辑验证
时序哈希链
同态加密
拜占庭容错
cloud-based train control
temporary speed restriction
logic verifi cation
temporal hash chain
homomorphic encryption
Byzantine fault tolerance
