This paper presents a method for designing a class of countermeasures for DPA attacks based on attenuation of current variations. In this class of countermeasures, designers aim at decreasing the dynamic current varia...This paper presents a method for designing a class of countermeasures for DPA attacks based on attenuation of current variations. In this class of countermeasures, designers aim at decreasing the dynamic current variations to reduce the information that can be extracted from the current consumption of secure microsystems. The proposed method is based on a novel formula that calculates the number of current traces required for a successful DPA attack using the characteristics of the microsystem current signal and the external noise of the measurement setup. The different stages of the proposed method are illustrated through designing an example current flattening circuit. Meanwhile validity and applicability of the proposed formula is verified by comparing theoretical results with those obtained experimentally for the example circuit. The proposed formula not only estimates the required level of attenuation for a target level of robustness defined by design requirements, it also predicts the effectiveness of a countermeasure using simulation results therefore dramatically reducing the time to design of secure microsystems.展开更多
文摘This paper presents a method for designing a class of countermeasures for DPA attacks based on attenuation of current variations. In this class of countermeasures, designers aim at decreasing the dynamic current variations to reduce the information that can be extracted from the current consumption of secure microsystems. The proposed method is based on a novel formula that calculates the number of current traces required for a successful DPA attack using the characteristics of the microsystem current signal and the external noise of the measurement setup. The different stages of the proposed method are illustrated through designing an example current flattening circuit. Meanwhile validity and applicability of the proposed formula is verified by comparing theoretical results with those obtained experimentally for the example circuit. The proposed formula not only estimates the required level of attenuation for a target level of robustness defined by design requirements, it also predicts the effectiveness of a countermeasure using simulation results therefore dramatically reducing the time to design of secure microsystems.
文摘人工养殖的雄性梅花鹿在发情期间攻击行为剧增,易造成鹿茸损伤,自动监测其攻击行为能为研究减少攻击行为提供重要依据。本文基于注意力机制和长短记忆序列研究了一种光流注意力网络(Optical flow attention attacking recognition network,OAAR),对梅花鹿的攻击、采食、躺卧、站立行为进行识别。OAAR网络包括前置网络、基础网络和时序网络,前置网络由LK光流算法(Lucas kanade optical flow algorithm)组成,用于提取RGB数据光流信息;基础网络中采用自注意力模块,将ResNet-152网络改造为ARNet152(Attention ResNet-152),用于将RGB、光流数据集经ARNet152提取特征后输入时序网络;时序网络采用添加注意力模块的长短记忆序列(Attention long short term network,ALST),并通过分类器输出行为得分和分类结果。视频数据集包括10942段,共310574帧,划分为攻击、采食、站立和躺卧4个大类,攻击行为又划分为撞击、脚踢和追逐3个小类;训练集、验证集和测试集比例为3∶1∶1。研究结果显示,OAAR模型在测试集上正确率为97.45%、召回率为97.46%、F1值为97.45%,ROC曲线中各类识别效果良好,特征嵌入图中各类行为特征区分度较高,各项结果均优于LSTM、双流I3D和双流ITSN网络,具有较好的泛化能力和抗干扰性。在本研究算法基础上集成的鹿只行为自动识别采集系统,为提高梅花鹿养殖生产管理水平和生产效率提供了技术基础。
