arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2311.09489cs.CR
收藏

MirrorNet:一种面向TEE的端侧DNN推理安全框架

MirrorNet: A TEE-Friendly Framework for Secure On-device DNN Inference

Ziyu Liu, Yukui Luo, Shijin Duan, Tong Zhou, Xiaolin Xu

AI总结:

MirrorNet利用TEE生成安全友好的DNN实现,通过轻量级镜像分支监控并修正输出,在保护模型机密性的同时实现低延迟,实验显示合法与非法使用精度差18.6%,硬件开销仅0.99%。

摘要:

深度神经网络(DNN)模型在边缘设备上用于实时推理已变得普遍。然而,它们容易受到模型提取攻击,需要保护。现有的防御方法要么无法完全保障模型机密性,要么导致显著的延迟问题。为克服这些挑战,本文提出MirrorNet,它利用可信执行环境(TEE)实现安全的端侧DNN推理。它为任何给定的DNN模型生成一个TEE友好的实现,以保护模型机密性,同时满足TEE严格的计算和存储约束。该框架由两个关键组件组成:主干模型(BackboneNet),存储于普通世界但推理精度较低,以及伴生部分监控器(CPM),一个存储于安全世界的轻量级镜像分支,用于保护模型机密性。在推理过程中,CPM监控来自BackboneNet的中间结果,并修正分类输出以达到更高精度。为增强灵活性,MirrorNet包含两个模块:CPM策略生成器,生成各种保护策略,以及性能模拟器,估计每种策略的性能并选择最优策略。大量实验证明了MirrorNet在提供安全保障的同时保持低计算延迟的有效性,使MirrorNet成为安全端侧DNN推理的实用且有前景的解决方案。评估中,MirrorNet在合法使用与非法使用之间实现了18.6%的精度差距,同时仅引入0.99%的硬件开销。

英文摘要:

Deep neural network (DNN) models have become prevalent in edge devices for real-time inference. However, they are vulnerable to model extraction attacks and require protection. Existing defense approaches either fail to fully safeguard model confidentiality or result in significant latency issues. To overcome these challenges, this paper presents MirrorNet, which leverages Trusted Execution Environment (TEE) to enable secure on-device DNN inference. It generates a TEE-friendly implementation for any given DNN model to protect the model confidentiality, while meeting the stringent computation and storage constraints of TEE. The framework consists of two key components: the backbone model (BackboneNet), which is stored in the normal world but achieves lower inference accuracy, and the Companion Partial Monitor (CPM), a lightweight mirrored branch stored in the secure world, preserving model confidentiality. During inference, the CPM monitors the intermediate results from the BackboneNet and rectifies the classification output to achieve higher accuracy. To enhance flexibility, MirrorNet incorporates two modules: the CPM Strategy Generator, which generates various protection strategies, and the Performance Emulator, which estimates the performance of each strategy and selects the most optimal one. Extensive experiments demonstrate the effectiveness of MirrorNet in providing security guarantees while maintaining low computation latency, making MirrorNet a practical and promising solution for secure on-device DNN inference. For the evaluation, MirrorNet can achieve a 18.6% accuracy gap between authenticated and illegal use, while only introducing 0.99% hardware overhead.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑