A New Evaluation Method On Anti-Attack Capability Of Cryptographic Chip

As a carrier of cryptographic algorithms,cryptographic chips are widely used in many aspects of life.However,the cryptographic chip will leak some physical information when running the encryption and decryption algorithm,and if the physical information is used by the attacker,the key information will be cracked.Due to the simple implementation process and low implementation cost of power consumption attacks,it has become a major threat to the security of cryptographic devices in recent years.The evaluation of the security performance of cryptographic chips has become particularly important..The success of the power attack is closely related to the number of samples collected.As the sample size increases,the attack success rate also increases.However,the sample size required for a successful chip attack can reach more than a million.In this paper,the statistical analysis of attack success rate under different sample sizes is carried out,and the relationship between attack success rate and sample size is obtained.Based on this,the relationship model between sample size and success rate is constructed.Then a cryptographic chip is proposed.Safety assessment method.This method can accurately predict the sample size required for successful attack when the sample attack is unsuccessful,so as to effectively evaluate the chip's anti-attack capability.In this paper,a SOC chip is used as the target board,and the DPA(Differential Power Analysis),which is commonly used in power consumption attacks,is used to analyze the power consumption.The power consumption data obtained by repeating the 8-bit repetition and 16-bit repeated key respectively The analysis was carried out to construct a model of the relationship between sample size and success rate.For the 32-bit duplicate key,3000 sets of power consumption data are collected into the model for analysis.The prediction result is that the key information can be successfully analyzed when the sample size is 20686.The experimental results show that when the sample size is 20000,the attack success rate has reached 98%,which is enough to crack the key.