• Convergence Security Journal
• /
• v.2 no.2
• /
• pp.165-177
• /
• 2002

IDS(Intrusion Detection System) evasion is a technology which uses vulnerability of IDS in order not to be detected by IDS. In this paper, at first, we classify IDS evasion technology. Second, we propose detection model of IDS evasion technology. Finally, we design and implement detection system of IDS evasion.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.907-917
• /
• 2023

Recently, a self-driving car have applied deep learning technology to advanced driver assistance system can provide convenience to drivers, but it is shown deep that learning technology is vulnerable to adversarial evasion attacks. In this paper, we performed five adversarial evasion attacks, including MI-FGSM(Momentum Iterative-Fast Gradient Sign Method), targeting the object detection algorithm YOLOv5 (You Only Look Once), and measured the object detection performance in terms of mAP(mean Average Precision). In particular, we present a method applying morphology operations for YOLO to detect objects normally by removing noise and extracting boundary. As a result of analyzing its performance through experiments, when an adversarial attack was performed, YOLO's mAP dropped by at least 7.9%. The YOLO applied our proposed method can detect objects up to 87.3% of mAP performance.

• Communications of the Korean Institute of Information Scientists and Engineers
• /
• v.36 no.2
• /
• pp.32-36
• /
• 2018

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.514-518
• /
• 2003

In this paper, we investigate three dimensional optimal evasive maneuver patterns for air-to-surface attack missiles against proportionally navigated anti-air defense missiles. Interception error of the defense missile can be generated by evasive maneuver of the attack missile during the time of flight for which the defense missile intercepts the attack missile. Time varying weighted sum of the inverse of these interception errors forms a performance index to be minimized. Direct parameter optimization technique using CFSQP is adopted to get the attack missile's optimal evasive maneuver patterns according to parameter changes of both the attack missile and the defense missile such as maneuver limit and time constant of autopilot approximated by the 1st order lag system. The overall shape of resultant optimal evasive maneuver to enhance the survivability of air-to-surface missiles against proportionally navigated anti-air missiles is a kind of deformed barrel roll.

• Journal of Information Processing Systems
• /
• v.12 no.3
• /
• pp.422-435
• /
• 2016

Despite the convenience brought by the advances in web and Internet technology, users are increasingly being exposed to the danger of various types of cyber attacks. In particular, recent studies have shown that today's cyber attacks usually occur on the web via malware distribution and the stealing of personal information. A drive-by download is a kind of web-based attack for malware distribution. Researchers have proposed various methods for detecting a drive-by download attack effectively. However, existing methods have limitations against recent evasion techniques, including JavaScript obfuscation, hiding, and dynamic code evaluation. In this paper, we propose an emulation-based malicious webpage detection method. Based on our study on the limitations of the existing methods and the state-of-the-art evasion techniques, we will introduce four features that can detect malware distribution networks and we applied them to the proposed method. Our performance evaluation using a URL scan engine provided by VirusTotal shows that the proposed method detects malicious webpages more precisely than existing solutions.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.12 no.3
• /
• pp.272-281
• /
• 2009

In this paper, three-dimensional optimal evasive maneuver patterns for air-to-surface attack missiles against proportionally navigated anti-air defense missiles were investigated. An interception error of the defense missile is produced by an evasive maneuver of the attack missile. It is assumed that the defense missiles are continuously launched during the flight of attack missile. The performance index to be minimized is then defined as the negative square integral of the interception errors. The direct parameter optimization technique based on SQP and a co-evolution method based on the augmented Lagrangian formulation are adopted to get the attack missile's optimal evasive maneuver patterns. The overall shape of the resultant optimal evasive maneuver is represented as a deformed barrel-roll.

• Journal of Korea Multimedia Society
• /
• v.25 no.6
• /
• pp.878-885
• /
• 2022

Research in the field of computer vision based on deep learning is being actively conducted. However, deep learning-based models have vulnerabilities in adversarial attacks that increase the model's misclassification rate by applying adversarial perturbation. In particular, in the case of FGSM, it is recognized as one of the effective attack methods because it is simple, fast and has a considerable attack success rate. Meanwhile, as one of the efforts to visualize deep learning models, Grad-CAM enables visual explanation of convolutional neural networks. In this paper, I propose a method to generate adversarial examples with high attack success rate by applying Grad-CAM to FGSM. The method chooses fixels, which are closely related to labels, by using Grad-CAM and add perturbations to the fixels intensively. The proposed method has a higher success rate than the FGSM model in the same perturbation for both targeted and untargeted examples. In addition, unlike FGSM, it has the advantage that the distribution of noise is not uniform, and when the success rate is increased by repeatedly applying noise, the attack is successful with fewer iterations.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.6
• /
• pp.1279-1290
• /
• 2021

As vehicle technology has grown, autonomous driving that does not require driver intervention has developed. Accordingly, CAN security, an network of in-vehicles, has also become important. CAN shows vulnerabilities in hacking attacks, and machine learning-based IDS is introduced to detect these attacks. However, despite its high accuracy, machine learning showed vulnerability against adversarial examples. In this paper, we propose a adversarial CAN frame generation method to avoid IDS by adding noise to feature and proceeding with feature selection and re-packet for physical attack of the vehicle. We check how well the adversarial CAN frame avoids IDS through experiments for each case that adversarial CAN frame generated by all feature modulation, modulation after feature selection, preprocessing after re-packet.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2003.11a
• /
• pp.1-4
• /
• 2003

최근에 침입 탐지 시스템은 네트워크 보안의 강화를 위해서 방화벽과 침입탐지 시스템 상호간의 연동으로 침입자의 연결 상태를 차단하는 방법도 개발되었다. 하지만 방화벽뿐만 아니라 침입탐지 시스템도 공격자에 의한 우회공격에 대해서는 아직 상당부분 방어할 수 없다. 또한 우회공격 탐지 모듈도 기존의 IDS와 Rule의 중복이 불가피하다. 본 논문은 취약점 진단 스크립트를 통해 IDS의 취약점 진단 후 IDS우회탐지공격 시스템의 Rule을 최적화 하여 우회공격을 효율적으로 탐지 해내는 시스템을 제안한다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.6
• /
• pp.1227-1236
• /
• 2021

Adversarial attack, one of the attacks on deep learning classification model, is attack that add indistinguishable perturbations to input data and cause deep learning classification model to misclassify the input data. There are various adversarial attack algorithms. Accordingly, many studies have been conducted to detect adversarial attack but few studies have been conducted to classify what adversarial attack algorithms to generate adversarial input. if adversarial attacks can be classified, more robust deep learning classification model can be established by analyzing differences between attacks. In this paper, we proposed a model that detects and classifies adversarial attacks by constructing a random forest classification model with input features extracted from a target deep learning model. In feature extraction, feature is extracted from a output value of hidden layer based on class predicted by the target deep learning model. Through Experiments the model proposed has shown 3.02% accuracy on clean data, 0.80% accuracy on adversarial data higher than the result of pre-existing studies and classify new adversarial attack that was not classified in pre-existing studies.