Detection And Analysis Of Side-Channel Leakage From FPGA Based Cryptographic Module

With the rapid development of the information industry,cryptographic modules have become increasingly widely used in the fields of politics,national defense,and finance.During the encryption process,it will leak side-channel power information,causing serious electromagnetic energy information security problems.The existing detection and evaluation technology about side-channel power leakage information from cryptographic modules lacks reliable standards,and evaluators cannot accurately assess the degree of security risk of the cryptographic module.Meanwhile the existing power analysis technology does not comprehensively consider the noise,imbalance and redundancy existing in the power trace,and the attackers cannot quickly and accurately crack the cryptographic module key.Therefore,it is particularly important and necessary to detect and analyze the side-channel leakage from cryptographic module.This paper concludes that the variance of the power traces come from the sensitive datadependent component(within-class variance)and noise(between-class variance)by analyzing the variance of the power trace.Evaluators estimate the minimum number of energy traces q required for a successful attack through multiple measurements of leakage power to evaluate the security performance of cryptographic modules;This paper integrates energy trace noise reduction,alignment and compression technology to proposes a joint preprocessing based power analysis method,which effectively improves the efficiency of power analysis on cryptographic module.The main work of the thesis is as follows:1.Build an FPGA-based AES-128 electromagnetic power leakage automatic acquisition system to quickly obtain the leaked power trace from the cryptographic module.This paper analyze the basic principles of AES encryption algorithm,study the mechanism of side-channel leakage from cryptographic modules and design a dedicated PCB for power analysis attack to automatically obtain the conduction and radiation leakage from the cryptographic module using 1? probes and near-field probes.2.In view of the lack of reliable evaluation indicators and low evaluation efficiency of sidechannel power leakage detection and evaluation,a side-channel information leakage risk evaluation strategy based on variance of power trace is proposed.This paper decomposes the variance of the traces of power consumption into variance of the sensitive variable and noise to estimate the signal-to-noise(SNR).According to the relationship between the success rate(SR)of the power analysis attack,SNR and conditional mutual information(conditional mutual information between the all power consumption and the power consumption from the sensitive variable given plaintext),the minimum number of traces is computable to evaluate the information leakage of the cryptographic chip.This method effectively estimate the minimum number of power traces when the SNR of power consumption is greater than 0.1.The result calculated by this method is consistent with the change trend of the simulation when the SNR is less than 0.1,which means that the result also evaluate the ability of cryptographic modules to resist power analysis attacks to a certain extent.The method improves the evaluation efficiency by at least 60000 times compared with the simulation of CPA.3.Aiming at the problems of noise,misalignment,and redundancy in the leakage energy traces of cryptographic modules,power analysis method based on joint preprocessing is proposed.Apply k-means++ algorithm to cluster the variance of the power trace,and calculate the variance threshold according to the obtained cluster center,remove redundant sampling points that are smaller than the threshold in the sampling interval,reduce computational complexity,and effectively improve attack efficiency;Combined energy trace alignment,compression and screening technologies,Apply DTW algorithm to align the leakage energy traces and reduce energy trace out-of-schedule;Use k-means++ algorithm to compress redundant energy traces to determine the range of attack targets;Use principal component analysis technology to filter energy traces with good signal-to-noise ratio,and perform related energy analysis attacks on the energy traces after screening.The results show that compared with the method without preprocessing technology,the number of power traces required for a successful attack is reduced by 54%,and the attack range is reduced by at least 2.5 times,which effectively improves attack efficiency.