8,938 research outputs found
Cho, Kwang Rae
No full text학위논문(석사)--아주대학교 정보통신대학원 :사이버보안,2020. 2사이버 공격은 DDoS, 홈페이지 위·변조, 악성코드 유포 공격 등 다양하게 이루어지고 있다. 그 중 악성코드 유포 공격은 집단 규모가 커지고 점차 조직화 되고 있으며 PC, 스마트폰 등에 대한 사이버 공격 피해 사례가 늘어나고 있다. 또한, 악성코드 유포 속도에 비해 탐지 속도가 뒤처지고 있어 효과적인 대응이 이루어지지 못하는 실정이다. 이에 따라 다양한 분야에서 악성코드 탐지를 위한 딥러닝 및 머신러닝 기법을 이용한 연구가 활발히 진행되고 있다. 그 중 CNN(Convolutional Neural Network)를 이용한 악성코드 분류의 정확도는 기존의 분류방법보다 정확도가 높은 것을 알 수 있다.
본 논문에서는 수집한 악성코드를 대상으로 CNN을 이용하여 바이트 코드 전체를 입력데이터로 하는 방법과 중요 섹션만을 추출하여 입력데이터로 사용한 방법을 실험을 통해 비교한 결과 중요 섹션만을 추출한 유형 분류의 정확도가 더 향상된 것을 확인하였다.제 1 장 서론 ········· ············1
제 1 절 연구 배경 및 목적 ··············1
제 2 절 연구의 구성 ·················4
제 2 장 머신러닝 및 딥러닝 기술 ············5
제 1 절 머신러닝(Machine Learning) ··········5
1. 지도 학습(Supervised Learning) ··········6
2. 비지도학습(Unsupervised Learning) ········8
3. 강화학습(ReinForcement Learning) ········8
제 2 절 딥 러닝(Deep Learning) ············9
1. 인공신경망(Artificial Neural Network) ·······9
2. 퍼셉트론(Perceptron) ··············10
제 3 절 합성곱 신경망(Convolutional Neural Network)····12
1. 합성곱 계층(Convolution Layer) ·········13
2. 풀링 계층(Pooling Layer) ············14
3. 활성화 함수··················14
제 3 장 기존 악성코드 탐지 및 분석 기술 ·······15
제 1 절 악성코드 ·················15
제 2 절 악성코드 분석 방법 ·············16
1) 자동화 분석(Fully-automated analysis) ······17
2) 정적 분석(Static Analysis) ············17
3) 동적 분석(Dynamic Analysis) ··········18
4) 상세 분석(Manual code reversing) ········18
제 3 절 딥러닝 기반 악성코드 연구 ··········19
제 4 장 제안하는 악성코드 식별 및 유형 분류 방법 ···21
제 1 절 실험 환경 ·················22
제 2 절 샘플 파일 ·················22
제 3 절 실험 절차 ················23
1) 악성코드 이미지 변환 ············23
2) 학습, 검증 데이터셋 구축 ··········26
3) 신경망 구축 ·················27
제 4 절 실험 결과 ·················29
제 5 장 결 론 ····················31
참고문헌 ·······················32Maste
Cho Hyung-Rae
No full text학위논문(석사)--아주대학교 정보통신전문대학원 :정보통신공학과,2009. 2이동 통신의 범위가 한 개의 단말에서 전체 네트워크로 이동하면서 Network Mobility(NEMO)에 연구가 진행 되었다. NEMO환경에서 Mobile IP-TV, Mobile Conferencing과 같은 응용을 제공하기 위해서는 멀티캐스트 지원 방식이 필요하지만, NEMO Basic Support Protocol(NBS)은 터널 중첩과 같은 문제를 가지게 된다. 또한 이동 통신 환경에서 빈번한 멀티캐스트 경로의 재설정 문제가 존재한다.
본 논문에서는 Proxy방식을 응용한 컴포넌트인 Intelligent NEMO Multicast Component(INMC)를 NEMO환경에 추가하여 터널 중첩 문제와 빈번한 경로 재설정 문제를 해결하였다. 새롭게 추가 된 컴포넌트에서는 액세스 라우터에 포함되며, 최상위 라우터들의 멀티캐스트 세션과 경로를 관리하게 된다. 새롭게 참여하는 단말과 라우터의 정보는 INMC에 업데이트 되어 기존의 정보와 함께 관리된다. 만약 이동 라우터가 이동을 하게 되어 요구하는 멀티캐스트 서비스가 INMC에서 관리되고 있다면 기존의 방식과는 달리 HA와 새로운 터널을 형성하지 않고 서비스가 가능해진다. HA와 새롭게 터널을 형성하지 않기 때문에 Bingding에 필요한 패킷(BU-BA)를 전송하지 않고 서비스가 가능해져 네트워크 자원 사용의 효율성을 가져오게 된다. 또한 멀티캐스트 핸드오버에 관여하는 새로운 알고리즘을 제시하여 동적인 핸드오버 전략을 소개한다. 또한 빈번한 이동으로 인한 멀티캐스트 경로 재설정에 관한 문제도 해결하였기 때문에 서비스 지연시간이 기존의 방식보다 줄어들게 된다. 위에서 제안한 방식을 기존의 NBS를 이용한 멀티캐스트 전송 기법과 비교하여 비용 절감 효과를 확인하였다.제 1 장 서론 .................................................................................... 1
제 2 장 관련 연구 ........................................................................... 4
제 1 절 NBS를 이용한 멀티캐스트 ..................................................... 4
제 2 절 NeMRI 프로토콜 ......................................................................... 7
제 3 절 HMIP ............................................................................................. 9
제 3 장 Proxy 방식을 응용한 멀티캐스트 전송기법 .................... 11
제 1 절 INMC의 구조 ............................................................................ 13
제 2 절 INMC를 이용한 멀티캐스트 전송과 핸드오버 ................. 15
제 4 장 성능 분석 ........................................................................... 27
제 1 절 시뮬레이션 환경 및 측정 기준............................................... 27
제 2 절 시뮬레이션 결과 ....................................................................... 28
제 5 장 결론 .................................................................................... 35
참고문헌 ........................................................................................... 37
Abstract .......................................................................................... 39Maste
Functional estimation of activity criticality indices and sensitivity analysis of expected project completion time
No full textFor a PERT network, a new method is developed for estimating the criticality index of activity i (ACI(i)) as a function of the expected duration of activity i (mu(i)) and for the sensitivity analysis of the expected project completion time (mu(T)) with respect to mu(i). The proposed method evaluates the frequency of activity i being on the critical path, and thereby its ACI(i) using Monte Carlo simulation or a Taguchi orthogonal array experiment at several values of pi, fits a logistic regression model for estimating ACI(i) as a function of mu(i), and then, using the estimated ACI(i) function, evaluates the amount of change in mu(T)-when mu(i) is changed by a given amount. Unlike the previous works, the proposed method models ACI(i) as a nonlinear (ie, logistic) function of mu(i), which can be used to estimate the amount of change in mu(T) for a variety of changes in mu(i). Computational results indicate that the performance of the proposed method is comparable to that of direct Monte Carlo simulation. Journal of the Operational Research Society (2004)
CADTM - A DATABASE TRANSACTION MANAGER FOR COORDINATING DESIGN ACTIVITIES IN CAD ENVIRONMENTS
No full textThis paper presents a new database transaction manager, CADTM, for coordinating design activities in CAD environments. Unlike the previous approaches for coordinating design activities, in which each designer faces complexity of a design activity, CADTM encapsulates the complexity of design activities from designers. This is achieved by the notion of transaction template which guides the designer not to produce incorrect design results due to misunderstanding of complex design activities, and by the notion of interleaving specification which enables the designer to cooperate with his/her group members in a consistent way. In CADTM, therefore, any designer does not need to know the details of the design activity, and also does not need to concern whether he can release intermediate results of a transaction
Development and evaluation of an uncertainty importance measure in fault tree analysis
No full textIn a fault tree analysis, an uncertainty importance measure is used to identify those basic events that significantly contribute to the uncertainty of the top-event probability. This paper defines an uncertainty importance measure of a basic event or of a group of basic events, and develops a two-stage procedure for experimentally evaluating the measure under the assumption that the probability of each basic event follows a lognormal distribution. The proposed method utilizes the Taguchi tolerance design technique with modifications. Then, the so-called contribution ratios which evaluate the main and/or interaction effects of the uncertainties of log-transformed basic-event probabilities on the uncertainty of the log-transformed top-event probability are calculated. The contribution ratios are used to estimate the defined uncertainty importance measure of a basic event or of a group of basic events. The proposed method consists of two stages for computational efficiency. In the first stage, the basic events with negligible effects on the uncertainty of the log-transformed top-event probability are screened out, and more detailed analyses are conducted in the second stage with a substantially smaller number of basic events. In addition, this paper presents an analysis method to quantify the percentage reduction in the uncertainty of the log-transformed top-event probability when the uncertainty of each basic-event probability is reduced. (C) 1997 Elsevier Science Limited
An uncertainty importance measure of activities in PERT networks
No full textIn a PERT network, the importance of each activity needs to be assessed in order to identify those that warrant more attention from the project management. For this purpose, two measures are frequently used: criticality index; and uncertainty importance measure. While the former is used to identify which activity's expected duration to decrease in order to decrease the expected project completion time, the latter can be used to identify those that deserve more attention in reducing the magnitude of the uncertainty (variability) in the project completion time (T). This paper defines an uncertainty-importance measure of an activity, or of a pair of activities, and develops a method for evaluating the defined measure under the assumption that the durations of activities are independently and symmetrically distributed. The method utilizes Taguchi tolerance design technique with modifications. First, PERT networks are classified into two types: Type A with a dominantly longer path than the others, and Type B without such a path. For a Type A network, T can be closely approximated by the sum of the duration of activities on the dominantly longer path, and the contribution of the uncertainty in each activity duration to the uncertainty in T can be analytically determined with ease. For a Type B network. however, such an analytic approach is not feasible. In addition, activity durations may have curvature and interaction effects on T. We therefore employ an appropriate experimental strategy for Type B networks, and calculate the so called contribution ratios which evaluate the main and/or interaction effects of activity durations on the variability of T. Contribution ratios are used to estimate the defined uncertainty importance measure of an activity or of a pair of activities. The proposed method is easy to use and, in particular, computationally efficient for large-sized PERT networks as compared to other existing methods which require Monte Carlo simulation
Evaluation of surficial stability for homogeneous slopes considering rainfall characteristics
No full textShallow slope failures in residual soil during periods of prolonged infiltration are commonly occurring in the world. This study examines an infinite slope analysis to estimate the influence of infiltration on surficial stability of slopes by the limit equilibrium method. An approximate method that accommodates the boundary condition of a uniform rainfall has been proposed to evaluate the likelihood of shallow slope failure that is induced by a particular rainfall event. The method based on the Mein and Larson model, which provides an explicit solution for preponding infiltration, has been applied to various types of soil having measured unsaturated hydraulic properties. To compare results with those obtained from the approximate method, a series of numerical analyses were carried out. According to the results, with the use of properly estimated input parameters, the approximate method was found to give results that compare reasonably well with those of more rigorous finite element analyses
Build or Rent? An Economic Analysis of Electronic Mall Competition and Roles of Cyber-mediaries
No full textWe present a stylished model of duopoly of e-tailing competition that integrates price and e-advertising decision. Under the significant investment to build a sites identity on the Cyberspace, firms may use some well-established cyber-mediaries rather than building their own new e-stores. From the Nash equilibria of this game, we derive the conditions of firms optimal strategies of choice of distribution. Nash equilibria can be characterised by the relative brand-identities to that of emall, and substitutibility of the products. The intensity of price competition is positively related to the emalls profit. However, the higher degree of price competition at the same site diminishes a producers incentive to join the mall. We also discuss the impacts of other forms of cyber-mediaries such as portal service with banner ads price comparison agents
Nanolithography using micro-scale mask enabled by hyperbolic metamaterial
No full textWe newly developed a nano-scale patterning method overcoming the diffraction limit of conventional photo-lithography technique by utilizing micro-scale aluminum mask-hyperbolic metamaterials hybrid structures, which is supported by our numerical simulation and experimental results
Instability of unsaturated soil slopes due to infiltration
No full textThis paper presents a procedure for calculating the safely factor for an unsaturated slope suffering from rainfall infiltration. The process of infiltration into a slope due to rainfall and its effect on soil slope behavior were examined using a two-dimensional finite element flow deformation coupled analysis program. The modified Mohr-Coulomb failure criterion was adopted for unsaturated soil strength to consider variations in strength due to the presence of matric suction. A safety factor was calculated based on the smoothed stress field obtained from finite element analysis, and an optimization technique was used to search for a critical slip surface. Some numerical examples are analyzed. The results illustrate a typical process of infiltration into unsaturated soil slopes and the effect of hydraulic conductivity on slope stability during rain-induced infiltration. (C) 2001 Elsevier Science Ltd. All rights reserved
- …
