To meet the security requirements of resource-constrained systems, a software PUF (SPUF) scheme based on the encryption algorithm is proposed using the timing violation to generate the chip’s fingerprint. In this work, the Camellia encryption algorithm is selected and implemented via a semi-custom design flow to get the deviation of circuit delay. And then, the delay time of critical path of the proposed circuit is generated through static timing analysis. Using the overclocking frequency, the Camellia encryption algorithm circuit operates at timing violation status. We can collect the wrong response data of the simulation result at different input signals. The wrong data caused by timing violation can be defined as output of SPUF, which is the chip’s fingerprint. To improve the randomness, the multi-round operation of the Camellia is used to process the SPUF data. The experimental results show that the uniqueness of SPUF is 50.01%, the SPUF data passes the NIST test, and no additional hardware overhead is required.

为了满足资源受限系统的安全性要求，提出了一种基于加密算法的软件PUF（SPUF）方案，该方案使用时序违规来生成芯片的指纹。在这项工作中，选择了山茶花加密算法，并通过半定制设计流程实现了该算法，以获取电路延迟的偏差。然后，通过静态时序分析，得出所提出电路关键路径的延迟时间。利用超频频率，山茶花加密算法电路在时序违规状态下运行。我们可以在不同的输入信号下收集模拟结果的错误响应数据。时序违规引起的错误数据可以定义为SPUF的输出，即芯片的指纹。为了提高随机性，茶花的多轮操作用于处理SPUF数据。实验结果表明，SPUF的唯一性为50.01％，SPUF数据通过了NIST测试，并且不需要额外的硬件开销。