• Convergence Security Journal
• /
• v.14 no.7
• /
• pp.53-58
• /
• 2014

MANET can quickly build a network because it is configured with only the mobile node and it is very popular today due to its various application range. However, MANET should solve vulnerable security problem that dynamic topology, limited resources of each nodes, and wireless communication by the frequent movement of nodes have. In this paper, we propose a domain-based distributed cooperative intrusion detection techniques that can perform accurate intrusion detection by reducing overhead. In the proposed intrusion detection techniques, the local detection and global detection is performed after network is divided into certain size. The local detection performs on all the nodes to detect abnormal behavior of the nodes and the global detection performs signature-based attack detection on gateway node. Signature DB managed by the gateway node accomplishes periodic update by configuring neighboring gateway node and honeynet and maintains the reliability of nodes in the domain by the trust management module. The excellent performance is confirmed through comparative experiments of a multi-layer cluster technique and proposed technique in order to confirm intrusion detection performance of the proposed technique.

• Convergence Security Journal
• /
• v.18 no.5_1
• /
• pp.45-51
• /
• 2018

As data utilization and importance becomes important, data-related accidents and damages are gradually increasing. Especially, insider threats are the most harmful threats. And these insider threats are difficult to detect by traditional security systems, so rule-based abnormal behavior detection method has been widely used. However, it has a lack of adapting flexibly to changes in new attacks and new environments. Therefore, in this paper, we propose an adaptive anomaly movement detection framework based on a statistical Markov model to detect insider threats in advance. This is designed to minimize false positive rate and false negative rate by adopting environment factors that directly influence the behavior, and learning data based on statistical Markov model. In the experimentation, the framework shows good performance with a high F2-score of 0.92 and suspicious behavior detection, which seen as a normal behavior usually. It is also extendable to detect various types of suspicious activities by applying multiple modeling algorithms based on statistical learning and environment factors.

• Journal of Internet Computing and Services
• /
• v.11 no.6
• /
• pp.73-86
• /
• 2010

In proportion to the rapid increase in the number of Web users, web attack techniques are also getting more sophisticated. Therefore, we need not only to detect Web attack based on the log analysis but also to extract web attack events from audit information such as Web firewall, Web IDS and system logs for detecting abnormal Web behaviors. In this paper, web attack detection system was designed and implemented based on integrated web audit data for detecting diverse web attack by generating integrated log information generated from W3C form of IIS log and web firewall/IDS log. The proposed system analyzes multiple web sessions and determines its correlation between the sessions and web attack efficiently. Therefore, proposed system has advantages on extracting the latest web attack events efficiently by designing and implementing the multiple web session and log correlation analysis actively.

• KIPS Transactions on Computer and Communication Systems
• /
• v.1 no.1
• /
• pp.55-60
• /
• 2012

Recent botnets are widely using the DNS services at the connection of C&C server in order to evade botnet's detection. It is necessary to study on DNS analysis in order to counteract anomaly-based technique using the DNS. This paper studies collection of DNS traffic for experimental data and supervised learning for DNS traffic-based malicious domain classification such as query of domain name corresponding to C&C server from zombies. Especially, this paper would aim to determine significant features of DNS-based classification system for malicious domain extraction by the Principal Component Analysis(PCA).

• Journal of the Korea Convergence Society
• /
• v.8 no.4
• /
• pp.19-24
• /
• 2017

The growing number of electronic financial transactions (e-banking) has entailed the rapid increase in security threats such as extortion and falsification of financial transaction data. Against such background, rigid security and countermeasures to hedge against such problems have risen as urgent tasks. Thus, this study aims to implement an improved case model by applying the Fraud Detection System (hereinafter, FDS) in a financial corporation 'A' using big data technique (e.g. the function to collect/store various types of typical/atypical financial transaction event data in real time regarding the external intrusion, outflow of internal data, and fraud financial transactions). As a result, There was reduction effect in terms of previous scenario detection target by minimizing false alarm via advanced scenario analysis. And further suggest the future direction of the enhanced FDS.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.14 no.6
• /
• pp.1097-1106
• /
• 2011

Tactical Information Communication Network(TICN), a concept-type integrated Military Communication system that enables precise command control and decision making, is designed to advance into high speed, large capacity, long distance wireless relay transmission. To support mobility in battlefield environments, the application of Ad-hoc networking technology to its wireless communication has been examined. Ad-hoc network works properly only if the participating nodes cooperate in routing and packet forwarding. However, if selfish nodes not forwarding packets of other nodes and malicious nodes making the false accusation are in the network, it is faced to many threats. Therefore, detection and management of these misbehaving nodes is necessary to make confident in Ad-hoc networks. To solve this problem, we propose an efficient intrusion detection technique to detect and manage those two types of attacks. The simulation-based performance analysis shows that our approach is highly effective and can reliably detect a multitude of misbehaving node.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10a
• /
• pp.730-732
• /
• 2001

최근 들어 버퍼오버플로우 취약성을 이용한 해킹 사례들이 늘어나고 있다. 버퍼오버플로우 공격을 탐지하는 방법은 크게 입력 데이터의 크기검사, 비정상적인 분기 금지, 비정상 행위 금지의 세가지 방식 중 하나를 취한다. 본 논문에서는 비정상적인 분기를 금지하는 방법을 살펴본 것이다. 기존의 방법은 부가적인 메모리를 필요로 하고, 컨트롤 플로우가 비정상적인 흐름을 찾기 위해 코드를 추가하고 실행함으로써 프로그램 실행시간의 저하를 단점으로 이야기할 수 있다. 본 논문에서는 부가적인 메모리 사용을 최소한으로 줄임으로 메모리 낭비를 저하시키고 실행시간에 컨트롤 플로우가 비정상적으로 흐르는 것을 막기 위한 작업들을 최소화 함으로서 기존의 방법보다 더 효율적인 방법을 제안하고자 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.132-135
• /
• 2021

최근 네트워크의 확장으로 인한 공격 벡터의 증가로 외부자뿐 아니라 내부자를 경계해야 할 필요성이 증가함에 따라, 이를 다룬 보안 모델인 제로트러스트 모델이 주목받고 있다. 이 논문에서는 reverse proxy 와 사용자 패턴 인식 AI 를 이용한 제로트러스트 아키텍처를 제시하며 제로트러스트의 구현 가능성을 보이고, 새롭고 효율적인 전처리 과정을 통해 효과적으로 사용자를 인증할 수 있음을 제시한다. 이를 위해 사용자별로 마우스 사용 패턴, 리소스 사용 패턴을 인식하는 딥러닝 모델을 설계하였다. 끝으로 제로트러스트 모델에서 사용자 패턴 인식의 활용 가능성과 확장성을 보인다.

• Journal of KIISE:Information Networking
• /
• v.31 no.6
• /
• pp.556-565
• /
• 2004

A Sandbox has been widely used to prevent damages caused by running of unknown malicious codes. It prevents damages by containing running environment of a program. There is a trade-off in using sandbox, between configurability and ease-of-use. MAPbox, an instance system of sandbox, had employed sandbox classification technique to satisfy both configurability and ease-of-use [1]. However, the configurability of MAPbox can be improved further. In this paper, we introduce a technique to attach dynamic class facility to MAPbox and implement MAPbox-advanced one. Newly generated class in our system has an access control with proper privileges. We show an example for improvements which denote our system have increased the configurability of MAPbox. It was determined as abnormal by MAPbox although is not. Our system could determine it as normal. We also show our techniques to overcome obstacles to implement the system.

• Journal of the Korea Convergence Society
• /
• v.9 no.7
• /
• pp.1-7
• /
• 2018

Feature selection, one of data preprocessing techniques, is one of major research areas in many applications dealing with large dataset. It has been used in pattern recognition, machine learning and data mining, and is now widely applied in a variety of fields such as text classification, image retrieval, intrusion detection and genome analysis. The proposed method is based on a genetic algorithm which is one of meta-heuristic algorithms. There are two methods of finding feature subsets: a filter method and a wrapper method. In this study, we use a wrapper method, which evaluates feature subsets using a real classifier, to find an optimal feature subset. The training dataset used in the experiment has a severe class imbalance and it is difficult to improve classification performance for rare classes. After preprocessing the training dataset with SMOTE, we select features and evaluate them with various machine learning algorithms.