• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2004년도 춘계종합학술대회
• /
• pp.391-394
• /
• 2004

인터넷의 급속한 확장으로 인해 네트워크 공격기법의 패러다임의 변화가 시작되었으며 새로울 공격 형태가 나타나고 있으나 대부분의 침입 탐지 기술은 오용 탐지 기술을 기반으로 하는 시스템이주를 이루고 있어 알려진 공격 유형만을 탐지하고, 새로운 공격에 능동적인 대응이 어려운 실정이다. 이에 새로운 공격 유형에 대한 탐지력을 높이기 위해 인체 면역 메커니즘을 적용하려는 시도들이 나타나고 있다. 본 논문에서는 데이터 마이닝 기법을 이용하여 네트워크 패킷에 대한 정상 행위 프로파일을 생성하고 생성된 프로파일을 자기공간화 하여 인체면역계의 자기, 비자기 구분기능을 이용해 자기 인식 알고리즘을 구현하여 이상행위를 탐지하고자 한다. 자기인식 알고리즘의 하나인 Negative Selection Algorithm을 기반으로 anomaly detector를 생성하여 자기공간을 모니터하여 변화를 감지하고 이상행위를 검출한다. DARPA Network Dataset을 이용하여 시뮬레이션을 수행하여 침입 탐지율을 통해 알고리즘의 유효성을 검증한다.

• Journal of Communications and Networks
• /
• 제10권1호
• /
• pp.89-97
• /
• 2008

Anomaly detection systems playa significant role in protection mechanism against attacks launched on a network. The greatest challenge in designing systems detecting anomalous exploits is defining what to measure. Effective yet simple, Shannon entropy metrics have been successfully used to detect specific types of malicious traffic in a number of commercially available IDS's. We believe that Renyi entropy measures can also adequately describe the characteristics of a network as a whole as well as detect abnormal traces in the observed traffic. In addition, Renyi entropy metrics might boost sensitivity of the methods when disambiguating certain anomalous patterns. In this paper we describe our efforts to understand how Renyi mutual information can be applied to anomaly detection as an offline computation. An initial analysis has been performed to determine how well fast spreading worms (Slammer, Code Red, and Welchia) can be detected using our technique. We use both synthetic and real data audits to illustrate the potentials of our method and provide a tentative explanation of the results.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.129-140
• /
• 2022

The effectiveness of system identification, damage detection, condition assessment and other structural analyses relies heavily on the accuracy and reliability of the measured data in structural health monitoring (SHM) systems. However, data anomalies often occur in SHM systems, leading to inaccurate and untrustworthy analysis results. Therefore, anomalies in the raw data should be detected and cleansed before further analysis. Previous studies on data anomaly detection mainly focused on just single type of data anomaly for denoising or removing outliers, meanwhile, the existing methods of detecting multiple data anomalies are usually time consuming. For these reasons, recognising multiple anomaly patterns for real-time alarm and analysis in field monitoring remains a challenge. Aiming to achieve an efficient and accurate detection for multi-type data anomalies for field SHM, this study proposes a pattern-recognition-based data anomaly detection method that mainly consists of three steps: the feature extraction from the long time-series data samples, the training of a pattern recognition neural network (PRNN) using the features and finally the detection of data anomalies. The feature extraction step remarkably reduces the time cost of the network training, making the detection process very fast. The performance of the proposed method is verified on the basis of the SHM data of two practical long-span bridges. Results indicate that the proposed method recognises multiple data anomalies with very high accuracy and low calculation cost, demonstrating its applicability in field monitoring.

• 인터넷정보학회논문지
• /
• 제22권1호
• /
• pp.13-22
• /
• 2021

최근 네트워크 환경에 대한 공격이 급속도로 고도화 및 지능화 되고 있기에, 기존의 시그니처 기반 침입탐지 시스템은 한계점이 명확해지고 있다. 지능형 지속 위협(Adavanced Persistent Threat; APT)과 같은 새로운 공격에 대해서 시그니처 패턴은 일반화 성능이 떨어지는 문제가 존재한다. 이러한 문제를 해결하기 위해 기계학습 기반의 침입 탐지 시스템에 대한 연구가 활발히 진행되고 있다. 하지만 실제 네트워크 환경에서 공격 샘플은 정상 샘플에 비해서 매우 적게 수집되어 클래스 불균형(Class Imbalance) 문제를 겪게 된다. 이러한 데이터로 지도 학습 기반의 이상 탐지 모델을 학습시킬 경우 정상 샘플에 편향된 결과를 가지게 된다. 본 논문에서는 이러한 불균형 문제를 해결하기 위해서 오토 인코더(Auto Encoder; AE)를 활용해 One-Class Anomaly Detection 을 수행하여 이를 극복한다. 실험은 NSL-KDD 데이터 셋을 통해 진행되었으며, 제안한 방법의 성능 평가를 위해 지도 학습된 모델들과 성능을 비교한다.

• 한국컴퓨터정보학회논문지
• /
• 제8권1호
• /
• pp.103-113
• /
• 2003

프로그램 행위 침입 탐지 기법은 데몬 프로그램이나 루트 권한으로 실행되는 프로그램이 발생시키는 시스템 호출들을 분석하고 프로파일을 구축하여 침입을 효과적으로 탐지한다 시스템 호출을 이용한 이상 탐지는 단지 그 프로세스가 이상(anomaly)임을 탐지할 뿐 그 프로세스에 의해 영향을 받는 여러 부분에 대해서는 탐지하지 못하는 문제점을 갖는다. 이러한 문제점을 개선하는 방법이 베이지안 확률값 이용하여 여러 프로세스의 시스템 호출간의 관계를 표현하고, 베이지안 네트워크를 이용한 어플리케이션의 행위 프로파일링에 의해 이상 탐지 정보를 제공한다. 본 논문은 여러 침입 탐지 모델들의 문제점들을 극복하면서 이상 침입 탐지를 효율적으로 수행할 수 있는 베이지안 네트워크를 이용한 침입 탐지 방법을 제안한다 행위의 전후 관계를 이용한 정상 행위를 간결하게 프로파일링하며, 변형되거나 새로운 행위에 대해서도 탐지가 가능하다. 제안한 정상행위 프로파일링 기법을 UNM 데이터를 이용하여 시뮬레이션하였다.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.77-91
• /
• 2022

Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, such as system identification and damage detection. Therefore, developing efficient techniques is quite essential to recognize the anomalies in monitoring data. In this study, several machine learning techniques are explored and implemented to detect and classify various types of data anomalies. A field dataset, which consists of one month long acceleration data obtained from a long-span cable-stayed bridge in China, is employed to examine the machine learning techniques for automated data anomaly detection. These techniques include the statistic-based pattern recognition network, spectrogram-based convolutional neural network, image-based time history convolutional neural network, image-based time-frequency hybrid convolution neural network (GoogLeNet), and proposed ensemble neural network model. The ensemble model deliberately combines different machine learning models to enhance anomaly classification performance. The results show that all these techniques can successfully detect and classify six types of data anomalies (i.e., missing, minor, outlier, square, trend, drift). Moreover, both image-based time history convolutional neural network and GoogLeNet are further investigated for the capability of autonomous online anomaly classification and found to effectively classify anomalies with decent performance. As seen in comparison with accuracy, the proposed ensemble neural network model outperforms the other three machine learning techniques. This study also evaluates the proposed ensemble neural network model to a blind test dataset. As found in the results, this ensemble model is effective for data anomaly detection and applicable for the signal characteristics changing over time.

• ETRI Journal
• /
• 제43권3호
• /
• pp.511-523
• /
• 2021

The World Health Organization provides guidelines for managing the particulate matter (PM) level because a higher PM level represents a threat to human health. To manage the PM level, a procedure for measuring the PM value is first needed. We use a PM sensor that collects the PM level by laser-based light scattering (LLS) method because it is more cost effective than a beta attenuation monitor-based sensor or tapered element oscillating microbalance-based sensor. However, an LLS-based sensor has a higher probability of malfunctioning than the higher cost sensors. In this paper, we regard the overall malfunctioning, including strange value collection or missing collection data as anomalies, and we aim to detect anomalies for the maintenance of PM measuring sensors. We propose a novel architecture for solving the above aim that we call the hypothesis pruning generative adversarial network (HP-GAN). Through comparative experiments, we achieve AUROC and AUPRC values of 0.948 and 0.967, respectively, in the detection of anomalies in LLS-based PM measuring sensors. We conclude that our HP-GAN is a cutting-edge model for anomaly detection.

• 정보보호학회논문지
• /
• 제13권2호
• /
• pp.99-106
• /
• 2003

네트워크 환경에서의 다양한 침입은 심각한 위험을 초래 할 수 있기 때문에 침입을 효과적으로 탐지하기 위해 데이터마이닝 기법을 발전시켜 왔다. 비정상행위 탐지 기술은 순수 데이터로 학습한 후, 비정상행위를 탐지하기 때문에 정교한 정상행위 패턴 생성이 필수적이다. 순수한 학습 데이터의 생성은 시간과 비용이 많이 드는 단점이 있다. 따라서 네트워크 상의 데이터에 대한 특징을 파악하는 것이 중요하다. 본 논문에서는 데이터마이닝의 연관규칙 및 클러스터링기법을 비정상행위 탐지에 적용하였고, 패킷내의 판정 요소에 정보이론 척도를 적용하여 불필요한 데이터를 필터링하는 방법을 제시하였다. 또한 가변길이 트랜잭션을 네트워크상의 분석 단위를 정의하는 기준으로 제시하여 행위 패턴 생성에 보다 묘사성이 높음을 보였다.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.63-75
• /
• 2022

The raw data collected by structural health monitoring (SHM) systems may suffer multiple patterns of anomalies, which pose a significant barrier for an automatic and accurate structural condition assessment. Therefore, the detection and classification of these anomalies is an essential pre-processing step for SHM systems. However, the heterogeneous data patterns, scarce anomalous samples and severe class imbalance make data anomaly detection difficult. In this regard, this study proposes a convolutional neural network-based data anomaly detection method. The time and frequency domains data are transferred as images and used as the input of the neural network for training. ResNet18 is adopted as the feature extractor to avoid training with massive labelled data. In addition, the focal loss function is adopted to soften the class imbalance-induced classification bias. The effectiveness of the proposed method is validated using acceleration data collected in a long-span cable-stayed bridge. The proposed approach detects and classifies data anomalies with high accuracy.

• ETRI Journal
• /
• 제38권6호
• /
• pp.1145-1152
• /
• 2016

Because the nodes in a wireless sensor network (WSN) are mobile and the network is highly dynamic, monitoring every node at all times is impractical. As a result, an intruder can attack the network easily, thus impairing the system. Hence, detecting anomalies in the network is very essential for handling efficient and safe communication. To overcome these issues, in this paper, we propose a rule-based anomaly detection technique using roaming honeypots. Initially, the honeypots are deployed in such a way that all nodes in the network are covered by at least one honeypot. Honeypots check every new connection by letting the centralized administrator collect the information regarding the new connection by slowing down the communication with the new node. Certain predefined rules are applied on the new node to make a decision regarding the anomality of the node. When the timer value of each honeypot expires, other sensor nodes are appointed as honeypots. Owing to this honeypot rotation, the intruder will not be able to track a honeypot to impair the network. Simulation results show that this technique can efficiently handle the anomaly detection in a WSN.