摘要
物理不可克隆函数(PUF,physically unclonable function)通过提取芯片制造过程中无法避免引入的工艺偏差,可产生具有随机性、唯一性和防篡改特性的特征密钥。通过对PUF电路结构和工作原理的研究,提出一种基于现场可编程门阵列(FPGA,field-programmable gate array)的新型强弱混合型PUF(SWPUF,strong and weak PUF)电路设计方案。该PUF可根据激励的汉明重量(HW,hamming weight)灵活地配置为强PUF和弱PUF两种拓扑结构,解决强/弱PUF分立实现的局限性。此外,利用异或去相关技术进一步优化输出密钥的统计特性。所提PUF采用Xilinx Artix-7 FPGA(28 nm工艺)实现,利用Matlab结合MicroBlaze微控制器构建内建自测试平台(self-built test platform)。实验结果表明,该PUF具有良好的随机性(96.98%)、唯一性(99.64%)和可靠性(常温常压下96.6%)。逻辑回归分析进一步显示,在HW较小的情况下所提SWPUF比传统的Arbiter-PUF具有更好的抗攻击能力,可广泛应用于信息安全领域,如密钥存储(针对弱PUF)和设备认证(针对强PUF)。
Physically unclonable function(PUF)can produce intrinsic keys with characteristics of randomness,uniqueness and tamper-proof by exploiting the process deviations which can not be avoided in the chip manufacturing process.A novel hybrid strong and weak PUF(SWPUF)circuit design based on field-programmable gate array(FPGA)was proposed after the investigation of the PUF circuit structures and principles.To address the limitation of designing strong-PUF and weak-PUF discretely,SWPUF could be configured into two topologies conveniently depending on the Hamming Weight(HW)of the challenges.In addition,the statistical characteristics of the responses could be further improved by a XOR-decorrelation technique.The proposed SWPUF was implemented on a Xilinx Artix-7 FPGA(28nm technology),and a self-built test platform was set up by using Matlab and MicroBlaze microcontroller.Experimental results show that the SWPUF has good performances of randomness(96.98%),uniqueness(99.64%)and reliability(96.6%).Logic register analysis also shows that the SWPUF has a better anti-attack capability than the traditional Arbiter-PUF in the case of with small HW,and can be used in the information security,such as key storage(especially to weak PUF)and device authentication(especially to strong PUF).
作者
连佳娜
汪鹏君
李刚
马雪娇
翟官宝
LIAN Jiana;WANG Pengjun;LI Gang;MA Xuejiao;ZHAI Guanbao(College of Electrical and Electronic Engineering,Wenzhou University,Wenzhou 325000,China;College of Computer Science and Artificial Intelligence,Wenzhou University,Wenzhou 325000,China;Oujiang College,Wenzhou University,Wenzhou 325000,China)
出处
《网络与信息安全学报》
2021年第2期94-103,共10页
Chinese Journal of Network and Information Security
基金
国家重点研发计划(2018YFB2202100)
国家自然科学基金(61874078,61904125)
温州市基础性科研项目(G20190006,G20190003)。
