• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권1호
• /
• pp.454-475
• /
• 2018

This paper presents a novel Android malware classification model planned to classify and categorize Android malicious code at Drebin dataset. The amount of malicious mobile application targeting Android based smartphones has increased rapidly. In this paper, Restricted Boltzmann Machine and Deep Belief Network are used to classify malware into families of Android application. A texture-fingerprint based approach is proposed to extract or detect the feature of malware content. A malware has a unique "image texture" in feature spatial relations. The method uses information on texture image extracted from malicious or benign code, which are mapped to uncompressed gray-scale according to the texture image-based approach. By studying and extracting the implicit features of the API call from a large number of training samples, we get the original dynamic activity features sets. In order to improve the accuracy of classification algorithm on the features selection, on the basis of which, it combines the implicit features of the texture image and API call in malicious code, to train Restricted Boltzmann Machine and Back Propagation. In an evaluation with different malware and benign samples, the experimental results suggest that the usability of this method---using Deep Belief Network to classify Android malware by their texture images and API calls, it detects more than 94% of the malware with few false alarms. Which is higher than shallow machine learning algorithm clearly.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2000년도 춘계학술대회논문집
• /
• pp.670-678
• /
• 2000

Loose parts monitoring system(LPMS) is one of the important monitoring systems for the safe and efficient operation of the nuclear reactor, since it is LPMS that can early detect loose parts which may cause a significant damage in facilities or components of the plant. Nuclear power plants in Korea have recently experienced several loose part alarms due to the metallic impact and it is expected that the frequency of the loose part will be increased along the aging of the plants. In this paper, the status of loose parts monitoring technologies and facilities in Korean nuclear power plants is presented for the establishment of LPMS installation plan in some nuclear reactors which are not yet equipped with LPMS. Sensor specification, location and mounting method for loose parts monitoring were reviewed. As a result, the location and the mounting method of the properly chosen sensor was recommended. Data acquisition algorithms and discriminating rules of loose part impact signals were also reviewed. Actual alarm cases occurred by true impact signal and false impact signal were stated here.

• 융합보안논문지
• /
• 제15권1호
• /
• pp.115-138
• /
• 2015

지식정보사회의 산업현장에서 필요로 하는 인재상은 지식과 실무를 겸비한 차별화된 전문직업인을 요구하지만, 교육기관을 통해 배출되는 인력들의 직무수행능력은 산업현장 요구에 미치지 못하여, 재교육을 위한 시간과 비용을 재투자하여야 하는 문제점이 있다. 이러한 기존 교육과정에 대한 한계와 문제점을 극복하고 산업현장에서 요구하는 양질의 인력을 공급하기 위한 교육과정 개발이 시급한 실정이다. 따라서 본 연구에서는 국가차원에서 추진하고 있는 국가직무능력표준(NCS) 개발 기법을 활용하여 경비분야 교육의 현장적합성을 제고하고, 산업체가 요구하는 실질적인 교육과정 개발을 제안한다.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2003년도 추계종합학술대회
• /
• pp.580-583
• /
• 2003

본 논문에서는 반복적인 움직임의 있는 경우, 강건하게 해결 할 수 있는 움직임 검출 알고리즘을 제시한다. 기존에 차 영상을 이용한 움직임 검출방법은 밝기나 잡음에는 어느 정도 강건하지만, 일정 영역에서 동작하는 물체의 반복적인 움직임에 대해서는 움직임으로 오 인식하는 문제점을 자주 발생시킨다. 이러한 문제점을 해결하기 위해 영상에서 반복적인 움직임은 특정 영역 상에서만 움직임의 발생된다는 특징을 이용해, 움직임의 가장 많이 발생한 영역을 움직임 영역으로 설정하고, 블록 정합(Block Matching) 시켜 계산된 평균절대오차(MAE)값을 가지고 문제를 해결하는 방법을 제시한다. 실험 결과 제안된 알고리즘은 다양한 반복적인 움직임에 대해 기존의 방법들에 비해 좋은 결과를 얻을 수 있었다.

• 한국산학기술학회논문지
• /
• 제13권1호
• /
• pp.382-389
• /
• 2012

위조된 보고서 공격은 무선 센서 네트워크에서 이벤트가 발생한 위치에 대한 송신 응답과 같은 거짓 경보를 야기하는 것뿐만 아니라 제한된 량의 에너지를 고갈시킨다. 본 논문에서는 위조된 보고서를 필터링하기 위해 확률적인 보안 필터링 기법(PFSS: Probabilistic Filtering method for Sensor network Security)을 제안한다. 제안 내용은 클러스터 헤드와 기지국과의 거리를 이용하여 기지국까지의 중간 클러스터 헤드가 검증 노드인지를 확률적으로 선택하여 보안 검증에 필요한 에너지를 줄이고, 보안 처리에 따른 핫 스팟 문제를 완화시킨다. 제안된 기법의 성능은 수식 분석과 실험을 통하여 분석하였으며, 이를 통하여 제안된 기법이 기존의 보안 검증 처리에 비해 효율적임을 알 수 있다.

• Journal of Communications and Networks
• /
• 제18권6호
• /
• pp.948-961
• /
• 2016

Internet protocol (IP) spoofing is a serious problem on the Internet. It is an attractive technique for adversaries who wish to amplify their network attacks and retain anonymity. Many approaches have been proposed to prevent IP spoofing attacks; however, they do not address a significant deployment issue, i.e., filtering inefficiency caused by a lack of deployment incentives for adopters. To defeat attacks effectively, one mechanism must be widely deployed on the network; however, the majority of the anti-spoofing mechanisms are unsuitable to solve the deployment issue by themselves. Each mechanism can work separately; however, their defensive power is considerably weak when insufficiently deployed. If we coordinate partially deployed mechanisms such that they work together, they demonstrate considerably superior performance by creating a synergy effect that overcomes their limited deployment. Therefore, we propose a universal anti-spoofing (UAS) mechanism that incorporates existing mechanisms to thwart IP spoofing attacks. In the proposed mechanism, intermediate routers utilize any existing anti-spoofing mechanism that can ascertain if a packet is spoofed and records this decision in the packet header. The edge routers of a victim network can estimate the forgery of a packet based on this information sent by the upstream routers. The results of experiments conducted with real Internet topologies indicate that UAS reduces false alarms up to 84.5% compared to the case where each mechanism operates individually.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권8호
• /
• pp.4021-4037
• /
• 2018

In network intrusion detection research, two characteristics are generally considered vital to building efficient intrusion detection systems (IDSs): an optimal feature selection technique and robust classification schemes. However, the emergence of sophisticated network attacks and the advent of big data concepts in intrusion detection domains require two more significant aspects to be addressed: employing an appropriate big data computing framework and utilizing a contemporary dataset to deal with ongoing advancements. As such, we present a comprehensive approach to building an efficient IDS with the aim of strengthening academic anomaly detection research in real-world operational environments. The proposed system has the following four characteristics: (i) it performs optimal feature selection using information gain and branch-and-bound algorithms; (ii) it employs machine learning techniques for classification, namely, Logistic Regression, Naïve Bayes, and Random Forest; (iii) it introduces bulk synchronous parallel processing to handle the computational requirements of large-scale networks; and (iv) it utilizes a real-time contemporary dataset generated by the Information Security Centre of Excellence at the University of Brunswick (ISCX-UNB) to validate its efficacy. Experimental analysis shows the effectiveness of the proposed framework, which is able to achieve high accuracy, low computational cost, and reduced false alarms.

• Journal of Power Electronics
• /
• 제16권3호
• /
• pp.1012-1024
• /
• 2016

This study deals with an on-line software fault detection, localization, and system reconfiguration method for electrical system drives composed of three-phase AC/DC/AC converters and three-phase permanent magnet synchronous machine (PMSM) drives. Current sensor failure (outage), speed/position sensor loss (disconnection), and damaged DC-link voltage sensor are considered faults. The occurrence of these faults in PMSM drive systems degrades system performance and affects the safety, maintenance, and service continuity of the electrical system drives. The proposed method is based on the monitoring signals of "abc" currents, DC-link voltage, and rotor speed/position using a measurement chain. The listed signals are analyzed and evaluated with the generated residuals and threshold values obtained from a Sliding Mode Current-Speed-DC-link Voltage Observer (SMCSVO) to acquire an on-line fault decision. The novelty of the method is the faults diagnosis algorithm that combines the use of SMCSVO and adaptive thresholds; thus, the number of false alarms is reduced, and the reliability and robustness of the fault detection system are guaranteed. Furthermore, the proposed algorithm's performance is experimentally analyzed and tested in real time using a dSPACE DS 1104 digital signal processor board.

• Journal of Power Electronics
• /
• 제11권6호
• /
• pp.801-810
• /
• 2011

An open circuit fault diagnostic method in IGBTs for the ac to dc converters used in microgrid applications is developed in this paper. An ac to dc converter is a key technology for microgrids in order to interface both distributed generation (DG) and renewable energy resources (RES). Also, highly reliable ac to dc converters are necessary to keep converters in continuous operation as long as possible during power switch fault conditions. Therefore, the proposed fault diagnostic method is developed to reduce the fault detection time and to avoid any other fault alarms because continuous operation is desired. The proposed diagnostic method is a combination of the absolute normalized dc current technique and the false alarm suppression algorithm to overcome the long fault detection time and fault alarm problems. The simulation and experimental results show that the developed fault diagnostic method can perform fault detection within about one cycle. The results illustrate that the reliability of an ac to dc converter interfaced with a microgrid can be improved by using the proposed fault diagnostic method.

• Journal of KIEE
• /
• 제11권1호
• /
• pp.69-77
• /
• 2001

This paper describes an image processing algorithm capable of recognizing road lanes by using a CDF(cumulative distribution function). The CDF is designed for the model function of road lanes. Based on the assumptions that there are no abrupt changes in the direction and location of road lanes and that the intensity of lane boundaries differs from that of the background, we formulated the CDF, which accumulates the edge magnitude for edge directions. The CDF has distinctive peak points at the vicinity of lane directions due to the directional and the positional continuities of a lane. To obtain lane-related information a scatter diagram was constructed by collecting edge pixels, of which the direction corresponds to the peak point of the CDF, then the principal axis-based line fitting was performed for the scatter diagram. Noises can cause many similar features to appear and to disappear in an image. Therefore, to reduce the noise effect a recursive estimator of the CDF was introduced, and also to prevent false alarms or miss detection a scene understanding index (DUI) was formulated by the statistical parameters of the CDF. The proposed algorithm has been implemented in real time on video data obtained from a test vehicle driven on a typical highway.