基于自比较方法的硬件木马检测技术

Hardware Trojan Detection Technology Based on the Self-Referencing Approach

硬件木马成为近年来硬件安全领域的一个新威胁。硬件木马可能在集成电路设计的各个阶段插入到电路中,一经流片,就不可修复更改,将对通信和安全等领域造成巨大损失。针对时序型硬件木马,提出一种不需要基准芯片的自比较硬件木马检测技术。即通过对电路添加相同重复的激励,比较对应激励在不同时间窗口的电流值,判断电路中是否含有木马。将高级加密标准(AES)加密电路在SMIC 0.18μm,0.13μm,65 nm,55 nm四种工艺库条件下进行仿真,并结合马氏距离处理实验数据,增加检测的区分度。仿真数据区分度提高了240.6倍,现场可编程门阵列(FPGA)实测数据增加了3.5倍。实验结果表明,基于自比较的硬件木马检测技术能够检测出比AES电路小2个数量级的硬件木马。

Hardware trojan has become a new threat in the hardware security field in recent years.It can be inserted into the circuits in different stages of the integrated circuit design,and can＇ t be restored after manufactured,which causes great damage to communication and security. Aiming at the sequential hardware trojan,a self-referencing method without the reference chip for detecting hardware trojans was proposed. By adding the same repetitive signals to the circuit,the current values of the corresponding signals at different time were compared to judge whether the circuit contains trojans. The advanced encryption standard（ AES） circuit was simulated under four process conditions of 0. 18 μm,0. 13 μm,65 nm and 55 nm of surface mount integrated circuit（ SMIC）. And the experimental data were analyzed with Mahalanobis distance to increase the discrimination of detection. The discrimination of simulation data was increased by 240. 6 times,and the experiment data of the field-programmable gate array（ FPGA） were increased by 3. 5 times. The results show that the hardware trojans which are 2 orders of magnitude smaller than the AES circuit can be detected with the hardware trojan detection technology based on the self-referencing approach.