ID 原文 译文
45026 然后利用径向基函数（RBF, radial basis function）神经网络分类器评估当前检查点的状态来判断软件的可信情况，并运用半加权马尔可夫模型预测下一个检查点的状态，达到对软件未来运行趋势的可信情况的评估。 Then the RBF neural network classifier evaluated the status of the current checkpoint to judge the trustworthiness of the software, and the semi weighted Markov model predicted the situation of the next checkpoint to evaluate the trustworthiness of future running trend of the software.
45027 实验结果证明了CBSI-TM 模型能够有效地预测软件未来运行趋势的可信情况，并验证了该模型具有更优的合理性和有效性。 The experimental results show that the CBSI-TM model can predict the future trusted status of the software effectively, and verify that the model is more reasonable and effective.
45028 在侧信道攻击中，作为抵抗一阶 DPA 攻击的对策，掩码策略是当前使用最为广泛的防御方式之一。 In side channel attack, the masking implementation is one of the most popular counter measures again first order DPA.
45029 目前，针对掩码策略，通常使用高阶 DPA 及高阶模板攻击等攻击方式。 Presently, high order DPA and high order template attack are often used to attack against masking countermeasures.
45030 但由于高阶 DPA 攻击的是多种信息的联合泄露，需要对多个位置的能耗进行交叉组合，导致其攻击效率低下。 High order DPA, however, targets joint leakage of multiple types of information and therefore needs cross combination of powers at corresponding positions, which is the root of the inefficiency of high order DPA.
45031 高阶模板攻击则需要在学习阶段了解每次加密中使用的随机掩码，攻击条件往往难以满足。 High order template attack, on the other hand, has to know the random mask in each encrypting at its learning phase, which is rarely satisfied for most adversaries.
45032 针对目前这些攻击方式的不足与局限性，有学习的高阶 DPA 采用神经网络建立能耗对无掩中间组合值的拟合模型， Be aware of these shortcomings and limitations, the algorithm of high order DPA with pro-filing used neural network to establish the model of fitting the combination of unmasked intermediate values.
45033 基于拟合无掩中间组合值与猜测无掩中间组合值的相关系数进行攻击。 Attacking was based on the correlation coefficient between the fitted combination intermediate value and the guessing combination intermediate value.
45034 这种方法消除了在学习阶段必须了解掩码的要求，同时避免了高阶 DPA 对能耗交叉组合的需求，降低了攻击条件，且提高了攻击的效率。 This method eliminated the requirement of knowing the masks at the learning phase of template at-tack and the requirement of cross combination of powers for high order DPA, and therefore lowered the requirement of learning as well as improved attacking efficiency.
45035 实验证实了该攻击算法的可行性和高效性。 Experiments have confirmed the feasibility of this algorithm, as well as its efficiency.