PUF-Based Anonymous Authentication Scheme for Hardware Devices and IPs in Edge Computing Environment
PUF-Based Anonymous Authentication Scheme for Hardware Devices and IPs in Edge Computing Environment
复制标题
DOI：
10.1109/access.2019.2925106
复制
发表时间：
2019
期刊：
影响因子：
3.9
通讯作者：
Jing Long;W. Liang;Kuan-Ching Li;Dafang Zhang;Mingdong Tang;Haibo Luo
中科院分区：
文献类型：
--
作者：
Jing Long;W. Liang;Kuan-Ching Li;Dafang Zhang;Mingdong Tang;Haibo Luo
随着边缘计算和物联网的快速发展，底层硬件设备的安全性越来越受到关注。现场可编程门阵列(现场可编程门阵列)作为一种理想的硬件解决方案，成为设计复杂系统的主流技术。所设计的模块被命名为知识产权(IP)核。本文综合考虑了边缘计算中硬件设备盗用和软件IP盗用的问题，提出了一种基于PUF的IP版权匿名认证方案。该方案利用双重物理不可克隆功能(PUF)认证模型。双方在认证中共同生成挑战，以避免重放攻击和对PUF电路的建模攻击。大大降低了身份验证的复杂性。此外，由于使用了双PUF认证模型，所以不需要在FPGA厂商的服务器上存储每个基于PUF的芯片的所有挑战响应对(CRP)。节省了系统资源，实现了更好的安全性。为了保护软件知识产权，IP核心供应商在交易前将版权信息和匿名买家身份信息插入到设计中。买家的匿名性确保了买家的利益。在可信设备厂商的参与下，可以根据提取的指纹追踪侵权行为。实验表明，与2-1 DAPUF和内置自调整PUF相比，该方案的资源开销分别减少了61.96%和31.61%。此外，PUF的稳定性为99.54%。实验结果表明，该方案具有良好的性能。
With rapid advances in edge computing and the Internet of Things, the security of low-layer hardware devices attract more and more attention. As an ideal hardware solution, field programmable gate array (FPGA) becomes a mainstream technology to design a complex system. The designed modules are named as intellectual property (IP) cores. In this paper, we consider both misappropriation of hardware devices and software IPs in edge computing and propose a PUF-based IP copyright anonymous authentication scheme. The scheme utilizes the double physical unclonable function (PUF) authentication model. Both the parties generate the challenge jointly in authentication to avoid replay attack and modeling attack on PUF circuit. The complexity of authentication is greatly reduced. Besides, the server of FPGA vendor is unnecessary to store all the challenge response pairs (CRPs) of each PUF-based chip due to the use of the double PUF authentication model. It saves the system resource and achieves better security. To protect software IP, IP core vendor inserts copyright information and anonymous buyer identity information into the design before trading. The anonymity of the buyer ensures the benefits of the buyer. With the participation of trustable device vendor, infringement behavior can be traced according to extracted fingerprints. The experiments show that the resource overhead of the proposed scheme is reduced by 61.96% and 31.61% by comparing with 2-1 DAPUF and built-in self-adjustable PUF. Besides, PUF stability is 99.54%. It demonstrates the good performance of the proposed scheme.