用轻量化技术让车载以太网入侵检测在低价设备上实时运行
Deep Learning-Based Intrusion Detection for Automotive Ethernet: Evaluating & Optimizing Fast Inference Techniques for Deployment on Low-Cost Platform
- 采用模型蒸馏与剪枝加速神经网络推理
- 在树莓派4上实现727微秒检测速度，AUCROC达0.9890
- 适合资源受限的车载系统部署，兼顾性能与效率
现代车辆连接性日益增强，汽车以太网作为车内通信的关键基础设施，面临流注入等安全威胁。基于深度学习的入侵检测系统（IDS）虽能应对，但通常需昂贵硬件支持实时运行。本文评估并应用模型蒸馏与剪枝等快速推理技术，实现低功耗平台上的实时部署。实验表明，在树莓派4上，该方法可将入侵检测时间缩短至727 μs，AUCROC达到0.9890，显著提升检测效率与可行性。
原文摘要 · Abstract (English)
Modern vehicles are increasingly connected, and in this context, automotive Ethernet is one of the technologies that promise to provide the necessary infrastructure for intra-vehicle communication. However, these systems are subject to attacks that can compromise safety, including flow injection attacks. Deep Learning-based Intrusion Detection Systems (IDS) are often designed to combat this problem, but they require expensive hardware to run in real time. In this work, we propose to evaluate and apply fast neural network inference techniques like Distilling and Prunning for deploying IDS models on low-cost platforms in real time. The results show that these techniques can achieve intrusion detection times of up to 727 μs using a Raspberry Pi 4, with AUCROC values of 0.9890.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。