Recent advances in deep learning‐based side‐channel analysis

As side‐channel analysis and machine learning algorithms share the same objective of classifying data, numerous studies have been proposed for adapting machine learning to side‐channel analysis. However, a drawback of machine learning algorithms is that their performance depends on human engineering. Therefore, recent studies in the field focus on exploiting deep learning algorithms, which can extract features automatically from data. In this study, we survey recent advances in deep learning‐based side‐channel analysis. In particular, we outline how deep learning is applied to side‐channel analysis, based on deep learning architectures and application methods. Furthermore, we describe its properties when using different architectures and application methods. Finally, we discuss our perspective on future research directions in this field.     

一种反求杂质浓度的数值方法——基于CV数据的逐点反求多次循环法

根据CV数据用数值模拟反求杂质浓度时,遇到了不确定性问题。本文认为这实际上是用CV数据反求杂质浓度时的非单值性问题。针对数值方法的非单值性问题,本文提出了逐点反求多次循环的算法。本算法已成功地应用于分布陡峻的高斯离子注入分布及矩形阶跃分布,反求的杂质浓度分布与理论杂质浓度分布符合得很好。     

基于硬件的内存trace工具 ——MTT的设计与实现

 本文提出了一种全新的获得访存trace的方式,并设计实现了基于硬件的零开销多平台实时访存Trace工具——MTT(Memory Trace Tool).详细介绍了MTT在采样配置、地址识别、trace输出等方面的设计细节,以及接收端配合MTT高效接收分析trace的流程,实现了一个通过MTT获得程序访存trace的完整方案.相比已有方法,MTT具有许多特点:(1)对程序透明;(2)零开销,无内存污染问题;(3)实时获取完整的全系统访存Trace;(4)可实时配置的多种在线Trace分析手段;(5)具有操作系统平台无关性.     

ERW钢管的数控仿形锯切技术概述

介绍了一种新型的钢管数控仿形锯切技术和设备,对几种类型的仿形锯的原理及特点进行了分析.该仿形锯切技术在国外已经应用多年,在国内也已引起很多钢管生产企业的重视,具有良好的发展前景.     

Modelling user preferences and configuring services in B3G devices

This paper discusses a management architecture for devices operating in heterogeneous environments, that enables access network selection through terminal-controlled, preference-based mechanisms. In this domain two problems are identified, mathematically formulated and solved: Intelligent Access Selection (IAS) and Modelling and Adaptation to User Preferences (MAUP). Their objective is to compute the optimal allocation of services to access networks and quality levels, and to dynamically determine user preferences according to the usage context, respectively. A greedy algorithm is proposed for the IAS problem, while the MAUP problem is handled through the construction of a Bayesian network that allows inference and learning of profile and usage patterns. Extensive simulation results of the proposed methods and algorithms are also presented.

多晶薄膜及表面层的XRD分析方法讨论

对多晶薄膜XRD分析的基本问题小角度衍射几何、非对称平行光束法和Seemann－Bohlin法的衍射几何强度计算，以及P－B法、S－B法对常规XRD分析的适应性进行了分析讨论。并对多晶薄膜XRD分析的当前进展、难点及发展作了分析。提出两种薄膜及表面层表征参量沿深度方向变化的情况下进行XRD深度分布分析的新方法，该方法具有真实深度尺度和定量的特点，并可用于界面分析。     

Fallacies of High‐Speed Hemodialysis

Chronic hemodialysis sessions, as developed in Seattle in the 1960s, were long procedures with minimal intra‐ and interdialytic symptoms. Financial and logistical pressures related to the overwhelming number of patients requiring hemodialysis created an incentive to shorten dialysis time to four, three, and even two hours per session in a thrice weekly schedule. This method spread rapidly, particularly in the United States, after the National Cooperative Dialysis Study suggested that time of dialysis is of minor importance as long as urea clearance multiplied by dialysis time and scaled to total body water (Kt/V_urea) equals 0.95–1.0. This number was later increased to 1.3, but the assumption remained unchanged that hemodialysis time is of minimal importance as long as it is compensated by increased urea clearance. Patients accepted short dialysis as a godsend, believing that it would not be detrimental to their well‐being and longevity. However, Kt/V_urea measures only removal of low molecular weight substances and does not consider removal of larger molecules. Besides, it does not correlate with the other important function of hemodialysis, namely ultrafiltration. Whereas patients with substantial residual renal function may tolerate short dialysis sessions, the patients with little or no urine output tolerate short dialyses poorly because the ultrafiltration rate at the same interdialytic weight gain is inversely proportional to dialysis time. Rapid ultrafiltration is associated with cramps, nausea, vomiting, headache, fatigue, hypotensive episodes during dialysis, and hangover after dialysis; patients remain fluid overloaded with subsequent poor blood pressure control, left ventricular hypertrophy, diastolic dysfunction, and high cardiovascular mortality. Short, high‐efficiency dialysis requires high blood flow, which increases demands on blood access. The classic wrist arteriovenous fistula, the access with the best longevity and lowest complication rates, provides “insufficient” blood flow and is replaced with an arteriovenous graft fistula or an intravenous catheter. Moreover, to achieve high blood flows, large diameter intravenous catheters are used; these fit veins “too tightly,” so predispose the patient to central‐vein thrombosis. Longer hemodialysis sessions (5–8 hrs, thrice weekly), as practiced in some centers, are associated with lower complication rates and better outcomes. Frequent dialyses (four or more sessions per week) provide better clinical results, but are associated with increased cost. It is my strong belief that a wide acceptance of longer, gentler dialysis sessions, even in a thrice weekly schedule, would improve overall hemodialysis results and decrease access complications, hospitalizations, and mortality, particularly in anuric patients.     

Depth profiling study of in situ CdCl2 treated CdTe/CdS heterostructure with glancing angle incidence X-ray diffraction

CdTe thin films have been deposited using spray pyrolysis technique without and with in situ CdCl₂ treatment. Scanning electron microscopy studies show enhanced grain growth in the presence of CdCl₂. Glancing angle incidence X-ray diffraction is used for the micro structural study of polycrystalline CdS/CdTe heterostructure at different depths by changing the incident angle. Spraying of CdCl₂ on CdS prior to CdTe deposition promotes S diffusion throughout CdTe film and also Te diffusion into CdS. Whereas spraying of CdCl₂ in between CdTe deposition prevents S diffusion partially and Te diffusion completely. There is an associated change in the microstress of the CdTe film at different layers. The films without CdCl₂ treatment show compressive microstress varying from −98 to −158 MPa with increasing incident angle. CdCl₂ spray during CdTe deposition shows compressive microstress, which varies from −98 MPa at the interface to −19 MPa near the surface and CdCl₂ spray prior to CdTe deposition leads to a mildly tensile stress, from +40 to +20 MPa, which is very close to the standard shear stress of 10 MPa for CdTe.     

Characterization of Arsenic segregation at Si/SiO2 interface by 3D atom probe tomography

Dopant loss due to the segregation and dopant pile up at the Si/SiO₂ interface are crucial phenomena in the scaling trend of MOSFET devices for the 22-nm technology node. Arsenic segregation and pile-up at the Si/SiO₂ interface have been studied by the atom probe tomography (APT) technique which allows the 3D observation and the chemical analyses of dopant distribution with the atomic scale resolution. Arsenic (10¹⁶ at/cm², 32 keV) was implanted in mono-crystalline silicon and then annealed at 900 °C for 6 h after a cleaning step and an oxide growing. The thickness of the segregation layer was determined at 2.3 nm containing 9.36 × 10¹⁴ at/cm² dose of segregated arsenic. Finally, the obtained arsenic segregated dose has been compared to the resistivity profile performed by spreading resistance profiling technique.     

An experimental investigation on the impact behavior of hybrid composite plates

In this experimental study, the impact behavior of hybrid composite plates has been investigated. The increasing impact energy was performed on two types of hybrid composite plates (glass–carbon/epoxy) until complete perforation of specimens. An energy profiling method, showing the relationship between impact energy and absorbed energy, was used together with load–deflection curves to determine the penetration and perforation thresholds of hybrid composites. The failure processes of damaged specimens for different impact energies were evaluated by comparing load–deflection curves and images of damaged samples taken from impacted sides and non-impacted sides. Cross-sections of damaged specimens were also inspected visually and discussed to assess the extent of damage, such as fiber fracture in layers, expansion of delaminations between adjacent layers. The perforation threshold of hybrid composite impacted from surface with carbon fibers was found approximately 30% higher than that of surface with glass fibers.