• Journal of Internet Computing and Services
• /
• v.24 no.4
• /
• pp.25-36
• /
• 2023

Today, as AI (Artificial Intelligence) technology is introduced in various fields, including security, the development of technology is accelerating. However, with the development of AI technology, attack techniques that cleverly bypass malicious behavior detection are also developing. In the classification process of AI models, an Adversarial attack has emerged that induces misclassification and a decrease in reliability through fine adjustment of input values. The attacks that will appear in the future are not new attacks created by an attacker but rather a method of avoiding the detection system by slightly modifying existing attacks, such as Adversarial attacks. Developing a robust model that can respond to these malware variants is necessary. In this paper, we propose two methods of generating Adversarial attacks as efficient Adversarial attack generation techniques for improving Robustness in AI models. The proposed technique is the XAI-based attack technique using the XAI technique and the Reference based attack through the model's decision boundary search. After that, a classification model was constructed through a malicious code dataset to compare performance with the PGD attack, one of the existing Adversarial attacks. In terms of generation speed, XAI-based attack, and reference-based attack take 0.35 seconds and 0.47 seconds, respectively, compared to the existing PGD attack, which takes 20 minutes, showing a very high speed, especially in the case of reference-based attack, 97.7%, which is higher than the existing PGD attack's generation rate of 75.5%. Therefore, the proposed technique enables more efficient Adversarial attacks and is expected to contribute to research to build a robust AI model in the future.

• Journal of National Security and Military Science
• /
• s.1
• /
• pp.191-229
• /
• 2003

Multi-sensor data fusion techniques combine evidences from multiple sensors in order to get more accurate and efficient meaningful information through several process levels that may not be possible from a single sensor alone. One of the most important parts in the data fusion system is the identification fusion, and it can be categorized into physical models, parametric classification and cognitive-based models, and parametric classification technique is usually used in multi-sensor data fusion system by its characteristic. In this paper, we propose a novel heuristic identification fusion method in which we adopt desirable properties from not only parametric classification technique but also cognitive-based models in order to meet the realtime processing requirements.

• Journal of Information Processing Systems
• /
• v.9 no.2
• /
• pp.247-258
• /
• 2013

A new secure network communication technique that has been designed for mobile wireless services, is presented in this paper. Its network services are mobile, distributed, seamless, and secure. We focus on the security of the scheme and achieve anonymity and reliability by using cryptographic techniques like blind signature and the electronic coin. The question we address in this paper is, "What is the best way to protect the privacy and anonymity of users of mobile wireless networks, especially in practical applications like e-commerce?" The new scheme is a flexible solution that answers this question. It efficiently protects user's privacy and anonymity in mobile wireless networks and supports various applications. It is employed to implement a secure e-auction as an example, in order to show its advantages in practical network applications.

• Electronics and Telecommunications Trends
• /
• v.35 no.6
• /
• pp.88-96
• /
• 2020

As the usability and value of personal information increase, the importance of privacy protection has increased. In Korea, the scope of the use of pseudonymized personal information has increased because of revisions to the law. In the past, security technologies were used to safely store and manage personal information, but now, security technologies focused on usability are needed to safely use personal information. In this paper, we look at issues related to the de-identification and re-identification of personal information. Moreover, we examine the standards and techniques related to the de-identification of personal information.

• Journal of information and communication convergence engineering
• /
• v.12 no.4
• /
• pp.252-256
• /
• 2014

Traditional block cipher Advanced Encryption Standard (AES) is widely used in the field of network security, but it has high overhead on each operation. In the 15th international workshop on information security applications, a novel lightweight and low-power encryption algorithm named low-power encryption algorithm (LEA) was released. This algorithm has certain useful features for hardware and software implementations, that is, simple addition, rotation, exclusive-or (ARX) operations, non-Substitute-BOX architecture, and 32-bit word size. In this study, we further improve the LEA encryptions for cloud computing. The Web-based implementations include JavaScript and assembly codes. Unlike normal implementation, JavaScript does not support unsigned integer and rotation operations; therefore, we present several techniques for resolving this issue. Furthermore, the proposed method yields a speed-optimized result and shows high performance enhancements. Each implementation is tested using various Web browsers, such as Google Chrome, Internet Explorer, and Mozilla Firefox, and on various devices including personal computers and mobile devices. These results extend the use of LEA encryption to any circumstance.

• Convergence Security Journal
• /
• v.10 no.2
• /
• pp.43-49
• /
• 2010

Nowadays, the occurrence of Malware is steadily increasing. The Malware is also becoming more intelligent, advanced and changing into various types. With the development of the information industry, the economic and monetary value of the information is going up and the damage due to the leaked information by the Malware is also increasing. This paper investigates the general usage of the User Agent in the HTTP Header, studies the Malware production techniques by transformation of the User-Agent information and suggests the technical and political counterplan against them.

• Journal of Advanced Navigation Technology
• /
• v.15 no.5
• /
• pp.887-892
• /
• 2011

Recently iPad has been released, so users interest in new portable device is increasing. As markets grow, experts are forecasting a increase of investigation about tablet PC. However iPad forensics is very difficult using existing smart phone forensic softwares. especially, those softwares can't analyze korean mobile application. This paper describes collecting/analyzing technique for iPad.

• Proceedings of the Korea Institutes of Information Security and Cryptology Conference
• /
• 2006.06a
• /
• pp.551-554
• /
• 2006

보안 침해사고시 수집된 디지털 증거를 법적 증거로써 제출하기 위해 신뢰성이 확보되어야한다. 이를 위한 디지털 증거 무결성 보증 기법들중 MDC를 사용한 디지털 증거 무결성 보증 기법은 MDC값을 공격자가 위조 변조 할 수 있다는 단점이 있다. 또 PKI를 이용한 공증 방식은 기존의 증거 수집 시스템을 수정해야 하는 단점과 새로 시스템을 도입하기 위한 비용이 많이 드는 단점이 있다. 따라서 본 논문에서 제안하는 디지털 증거 보호 기법은 Diffie-Hellman(DH) 키 교환 알고리즘을 이용하여 생성된 비밀키와 디지털 증거 수집 대상 시스템(Collecotion System-CS)의 하드웨어 정보($HW_{CS}$)로 디지털 증거(D), 디지털 증거에 대한 해쉬값(H(D))과 타임스탬프(Time)를 암호화해서 디지털 증거에 대한 기밀성, 인증 및 무결성을 보증하는 기법을 제안하였다.

• Proceedings of the Korea Institutes of Information Security and Cryptology Conference
• /
• 2006.06a
• /
• pp.291-294
• /
• 2006

기존의 DRM 보호기법은 특정 콘텐츠 포맷에 일부 제한적인 보호방식의 특성이 있기 때문에 현재 모든 디지털 콘텐츠에 대해서 범용적으로 DRM 보호기법을 적용할 수 없는 문제점이 있으므로, 아직도 많은 종류의 디지털 콘텐츠는 DRM 보호기능을 제공받지 못하고 있다. 또한, 기존의 DRM 보호기법은 암호화 알고리즘을 기반으로 많은 양의 데이터에 대해서 암/복호화를 수행하기 때문에 속도가 많이 소요되는 단점이 있다. 본 논문에서는 범용적인 DRM 보호를 위해 콘텐츠 헤더의 부분 암호화 방식과 분할된 콘텐츠에 대한 순열 재조합을 이용한 방법으로 모든 디지털 콘텐츠에 대해서 범용적으로 DRM 보호기능을 제공하는 기법을 제안하였다. 또한 이 기법은 콘텐츠의 암/복호화 수행 속도를 향상 시키고 동시에 기존의 보안성은 유지하는 기능을 포함하고 있으므로 다양한 형식의 디지털 콘텐츠에 대하여 유연하게 적용이 가능하며, 다양한 비즈니스 환경에 확장성 있는 통합 가능하다.

• Journal of Advanced Navigation Technology
• /
• v.13 no.4
• /
• pp.600-609
• /
• 2009

We provide indifferentiable security analyses of pfMD, MDP, WPH, EMD, NI and CS hash domain extensions and their truncated versions. Unlike previous analytic techniques, the analytic technique considered in this paper is simple and easy. Moreover, the analytic technique can be generally applied to any types of hash domain extensions. That means that the technique can be used as an analyzing tool for any new developed hash function.