• 한국정보통신학회논문지
• /
• 제25권1호
• /
• pp.69-74
• /
• 2021

코로나19의 영향으로 온라인 활동이 늘어나면서 인터넷 접속량도 늘어나고 있다. 하지만 악의적인 사용자에 의해서 네트워크 공격도 다양해지고 있으며 그중에서 DDoS 공격은 해마다 증가하는 추세이다. 이러한 공격은 침입 탐지 시스템에 의해서 탐지되며 조기에 차단할 수 있다. 침입 탐지 알고리즘을 검증하기 위해 다양한 데이터 세트를 이용하고 있으나 본 논문에서는 최신 트래픽 데이터 세트인 CICIDS2017를 이용한다. 의사 결정 트리를 이용하여 DDoS 공격 트래픽을 분석하였다. 중요도가 높은 결정적인 속성(Feature)을 찾아서 해당 속성에 대해서만 의사 결정 트리를 진행하여 정확도를 확인하였다. 그리고 위양성 및 위음성 트래픽의 내용을 분석하였다. 그 결과 하나의 속성은 98%, 두 가지 속성은 99.8%의 정확도를 각각 나타냈다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.2168-2187
• /
• 2021

The smart grid replaces the traditional power structure with information inventiveness that contributes to a new physical structure. In such a field, malicious information injection can potentially lead to extreme results. Incorrect, FDI attacks will never be identified by typical residual techniques for false data identification. Most of the work on the detection of FDI attacks is based on the linearized power system model DC and does not detect attacks from the AC model. Also, the overwhelming majority of current FDIA recognition approaches focus on FDIA, whilst significant injection location data cannot be achieved. Building on the continuous developments in deep learning, we propose a Deep Learning based Locational Detection technique to continuously recognize the specific areas of FDIA. In the development area solver gap happiness is a False Data Detector (FDD) that incorporates a Convolutional Neural Network (CNN). The FDD is established enough to catch the fake information. As a multi-label classifier, the following CNN is utilized to evaluate the irregularity and cooccurrence dependency of power flow calculations due to the possible attacks. There are no earlier statistical assumptions in the architecture proposed, as they are "model-free." It is also "cost-accommodating" since it does not alter the current FDD framework and it is only several microseconds on a household computer during the identification procedure. We have shown that ANN-MLP, SVM-RBF, and CNN can conduct locational detection under different noise and attack circumstances through broad experience in IEEE 14, 30, 57, and 118 bus systems. Moreover, the multi-name classification method used successfully improves the precision of the present identification.

• 정보보호학회논문지
• /
• 제22권4호
• /
• pp.819-825
• /
• 2012

최근 대다수 피싱사이트는 원시사이트(피싱사이트가 사칭하는 기관의 공식 웹사이트)와 유사하게 보이기 위해 원시사이트의 이미지, 게시글 등 컨텐츠를 링크하여 화면에 표시한다. 본 논문은 이러한 유형의 피싱사이트에 사용자가 접속하는 경우 피싱사이트의 URL이 HTTP referer 헤더필드를 통해 원시사이트로 유입되는 특성을 이용하여 피싱사이트를 실시간 탐지하는 시스템을 제안한다. 제안된 시스템은 원시사이트에 유입된 HTTP 트래픽을 아웃오브패스 (out-of-path) 방식으로 수집하여 분석함으로써 홈페이지 실운영 환경에 대한 영향을 최소화하였으며, 원시사이트를 참조한 웹 사이트의 URL에 대해 휴리스틱 분석을 실시함으로써 피싱사이트를 실시간으로 탐지할 수 있도록 설계하였다. 제안된 시스템을 피싱사이트 표적이 되고 있는 국내 모 기관 홈페이지에 적용한 결과 6일 동안 40개의 피싱사이트를 탐지하였다.

• 한국ITS학회 논문지
• /
• 제19권6호
• /
• pp.26-36
• /
• 2020

보행자도로의 서비스 수준을 판단하는 평가지표인 보행 교통류율, 점유공간, 밀도, 속도 등의 정량적 자료를 산출하는 것은 많은 시간 소요와 경제적 부담으로 작용한다. 따라서 적절한 보행자 정책 수립을 위한 대체 대안으로 보행자의 미시적 경로 파악과 이를 활용한 서비스 수준 판단기술이 요구된다. 본 연구에서는 지하철역인 역삼역을 중심으로 미시적 교통류 분석 프로그램인 VISSIM을 이용하여 네트워크를 구축하고 정산 및 검증을 통해 신뢰성을 확보하였다. 이를 통해 도로용량편람 상의 속도를 활용하여 보행자 도로의 서비스 수준을 판단해 보았다.

• 한국ITS학회 논문지
• /
• 제20권1호
• /
• pp.146-159
• /
• 2021

본 연구는 고령운전자 운전특성 파악을 위한 실증적 분석의 일환으로 차량 시뮬레이터를 이용하여 고령운전자의 주행특성을 파악하고자 하였다. 이를 위해 기존연구 고찰에서 고령운전자의 주행특성 요소를 파악한 후, 시뮬레이션 환경을 구축하고 일반 운전자와의 주행 비교를 통해 고령자 운전특성을 명확히 도출하고자 하였다. 실험 결과 고령운전자는 주행속도, 운전조작(브레이크, 스로틀, 스티어링 작동) 등 차로편측을 제외한 모든 항목에서 일반 운전자와는 다른 주행특성을 보이는 사실을 확인하였으며, 주요 특성으로 차로변경 시 속도유지 및 적정 차간간격 확보가 어렵고, 돌발 상황 발생 시 정지거리와 소요시간이 더 많이 필요한 것으로 확인되었다. 결과적으로 고령운전자 주행거동의 일의적 특성을 입증할 수 있었고, 도로설계, 고령자 관련 교통정책 수립 등에 유효한 지침을 제공할 수 있을 것으로 기대된다.

• 한국의류산업학회지
• /
• 제24권5호
• /
• pp.593-604
• /
• 2022

The purpose of this study is to analyze the flow of change in perception of metaverse before and after COVID-19 through big data analysis. This research method used Textom to collect all data, including metaverse for two years before COVID-19 (2018.1.1~2019.11.30) and after COVID-19 outbreak (2020.1.11~2021.12.31), and the collection channels were selected by Naver and Google. The collected data were text mining, and word frequency, TF-IDF, word cloud, network analysis, and emotional analysis were conducted. As a result of the analysis, first, hotels, weddings, and glades were commonly extracted as social issues related to metaverse before and after COVID-19, and keywords such as robots and launches were derived, so the frequency of keywords related to hotels and weddings was high. Second, the association of the pre-COVID-19 metaverse keywords was platform-oriented, content-oriented, economic-oriented, and online promotion-oriented, and post-COVID-19 clusters were event-oriented, ontact sales-oriented, stock-oriented, and new businesses. Third, positive keywords such as likes, interest, and joy before COVID-19 were high, and positive keywords such as likes, joy, and interest after COVID-19. In conclusion, through this study, it was found that metaverse has firmly established itself as a new platform business model that can be used in various fields such as tourism, travel, festivals, and education using smart technology and metaverse.

• Nuclear Engineering and Technology
• /
• 제53권12호
• /
• pp.3879-3891
• /
• 2021

A heat pipe residual heat removal system is proposed to be incorporated into the reactor driven subcritical (RDS) facility, which has been proposed by MIT Nuclear Reactor Laboratory for testing and demonstrating the Fluoride-salt-cooled High-temperature Reactor (FHR). It aims to reduce the risk of the system operation after the shutdown of the facility. One of the main components of the system is an air-cooled heat pipe heat exchanger. The alkali-metal high-temperature heat pipe was designed to meet the operation temperature and residual heat removal requirement of the facility. The heat pipe model developed in the previous work was adopted to simulate the designed heat pipe and assess the heat transport capability. 3D numerical simulation of the subcritical facility active zone was performed by the commercial CFD software STAR CCM + to investigate the operation characteristics of this proposed system. The thermal resistance network of the heat pipe was built and incorporated into the CFD model. The nominal condition, partial loss of air flow accident and partial heat pipe failure accident were simulated and analyzed. The results show that the residual heat removal system can provide sufficient cooling of the subcritical facility with a remarkable safety margin. The heat pipe can work under the recommended operation temperature range and the heat flux is below all thermal limits. The facility peak temperature is also lower than the safety limits.

• KEPCO Journal on Electric Power and Energy
• /
• 제8권1호
• /
• pp.15-19
• /
• 2022

The radial distribution systems (RDS) commonly used around the world has the following disadvantages. First, when the DL is operated on a radial system, the line utilization rate is usually kept low. Second, if a fault occurs in the radial DL, a power outage of 3 to 5 minutes is occurring depending on the operator's proficiency and fault situation until the fault section is separated and the normal section is replaced. To solve this problem, Various methods have been proposed at domestic and foreign to solve this problem, and in Korea, research is underway on the advanced system of operating multiple linked DL always. A system that is electrically linked always, and that is built to enable high-speed communication during the protection coordination is named networked distribution system (NDS). Because the load shares the DL, the line utilization rate can be improved, and even if the line faults, the normal section does not need to be cut off, so the normal section does not experience a power outage. However, since it is impossible to predict in which direction the fault current will flow when a failure occurs in the NDS, a communication-based protection coordination is used, but there is no backup protection coordination method in case of communication failure. Therefore, in this paper, we propose a protective cooperation method to apply as a backup method when communication fails in NDS. The new method is to change TCC by location of CB using voltage drop in case of fault.

• KEPCO Journal on Electric Power and Energy
• /
• 제8권1호
• /
• pp.43-48
• /
• 2022

Although the distribution system has been structured as complicated as a mesh in the past, the connection points for each line are always kept open, so that it is operated as a radial distribution system (RDS). For RDS, the line utilization rate is determined according to the maximum load on the line, and the utilization rate is usually kept low. In addition, when a fault occurs in the RDS, a power outage of about 3 to 5 minutes occurs until the fault section is separated, and the healthy section is transferred to another line. To improve the disadvantages of the RDS, research on the construction of a networked distribution system (NDS) that linking multiple lines is in progress. Compared to the RDS, the NDS has advantages such as increased facility utilization, load leveling, self-healing, increased capacity connected to distributed generator, and resolution of terminal voltage drop. However, when a fault occurs in the network distribution system, fault current can flow in from all connected lines, and the direction of fault current varies depending on the fault point, so a high-precision fault current direction determination method and high-speed communication are required. Therefore, in this paper, we propose an accurate fault current direction determination method by comparing the peak value polarity of the fault current in the event of a fault, and a communication-based protection coordination method using this method.

• 한국농촌건축학회논문집
• /
• 제24권3호
• /
• pp.47-57
• /
• 2022

This study attempted to grasp the status and perception of Uljin Geumgangsong by grasping mass media issues and user perception using big data, and to present basic data when constructing monitoring using user perception by examining the establishment relationship of agenda setting from a time-series perspective. The results of collecting and analyzing text data that can identify mass media and visitor awareness are as follows. First, both mass media and visitor keywords were related to the importance of the value and meaning of Uljin Geumgangsong. Second, in the case of the connection network, Geumgang Pine Agriculture was centered, but in the case of difference in perception between mass media and visitors, such results were derived due to the object of interest. Third, in the case of the connection relationship structure, the connection strength was strong because there were many overlapping contents of mass media. Fourth, as a result of the centrality analysis, both mass media and visitor-aware keywords were positively recognized as spaces created and maintained through institutional support, and objective perception could be grasped by finding hidden keywords. Fifth, as a result of time series analysis, it was possible to grasp the flow through the issue keywords that appeared by period, and unlike the past, it was recognized as a place for tourism and travel. Finally, as a result of examining whether the agenda setting is consistent, there is a mass media influence, so it is thought that more diverse and more information and publicity are needed by utilizing it.