• 정보보호학회논문지
• /
• 제33권3호
• /
• pp.411-425
• /
• 2023

오늘날 AI(Artificial Intelligence) 기술은 다양한 분야에서 활용 목적에 맞게 분류, 회기 작업을 수행하며 광범위하게 활용되고 있으며, 연구 또한 활발하게 진행 중인 분야이다. 특히 보안 분야에서는 예기치 않는 위협을 탐지해야 하며, 모델 훈련과정에 알려진 위협 정보를 추가하지 않아도 위협을 탐지할 수 있는 비 지도학습 기반의 이상 탐지 기법이 유망한 방법이다. 하지만 AI 판단에 대한 해석 가능성을 제공하는 선행 연구 대부분은 지도학습을 대상으로 설계되었기에 학습 방법이 근본적으로 다른 비 지도학습 모델에 적용하기는 어려우며, Vision 중심의 AI 매커니즘 해석연구들은 이미지로 표현되지 않는 보안 분야에 적용하기에 적합하지 않다. 따라서 본 논문에서는 침해공격의 원본인 최적화 Reference를 탐색하고 이와 비교함으로써 탐지된 이상에 대한 해석 가능성을 제공하는 기법을 활용한다. 본 논문에서는 산출된 Reference를 기반으로 실존 데이터에서 가장 가까운 데이터를 탐색하는 로직을 추가 제안함으로써 실존 데이터를 기반으로 이상 징후에 대한 더욱 직관적인 해석을 제공하고 보안 분야에서의 효과적인 이상 탐지모델 활용을 도모하고자 한다.

• 한국IT서비스학회지
• /
• 제11권sup호
• /
• pp.51-60
• /
• 2012

In this paper, we provide the analysis of device model and method between device models for compatible M&S V&V of the NetSPIN. First of all, we analysis features, structure, and classification of the NetSPIN. The second, as a part of reliable V&V process, we analysis network system modeling process, correlation between device modeling process for M&S of the NetSPIN. The third, we suggest making a kind of pool of reference model and module of devices for the increase factor of reuse between device model. We also, at the point view of M&S V&V, conclude that there is the validity of the fidelity in device modeling process. Through the analysis of the NetSPIN device model and suggestion of the method for higher compatibility between device modes, the development process of device model be clearly understood. Also we present the effective method of the development for reliable device mode as the point of V&V.

• 음성과학
• /
• 제15권2호
• /
• pp.55-65
• /
• 2008

Recently, the multi-model based speech recognizer has been used quite successfully for noisy speech recognition. For the selection of the reference HMM (hidden Markov model) which best matches the noise type and SNR (signal to noise ratio) of the input testing speech, the estimation of the SNR value using the VAD (voice activity detection) algorithm and the classification of the noise type based on the GMM (Gaussian mixture model) have been done separately in the multi-model framework. As the SNR estimation process is vulnerable to errors, we propose an efficient method which can classify simultaneously the SNR values and noise types. The KL (Kullback-Leibler) distance between the single Gaussian distributions for the noise signal during the training and testing is utilized for the classification. The recognition experiments have been done on the Aurora 2 database showing the usefulness of the model compensation method in the multi-model based speech recognizer. We could also see that further performance improvement was achievable by combining the probability density function of the MCT (multi-condition training) with that of the reference HMM compensated by the D-JA (data-driven Jacobian adaptation) in the multi-model based speech recognizer.

• 한국도서관정보학회지
• /
• 제48권4호
• /
• pp.129-151
• /
• 2017

현재 디지털 아카이빙에 널리 적용되고 있는 정보 모델링에서는 기술사항 생성을 위한 개념적인 프로세스만을 정의하고 있으며, 이를 정보자원의 보존 혹은 보존을 위한 메타데이터의 생성에 실제적으로 적용하는데 있어서는 한계를보이고 있다. 이에 본 연구에서는 디지털 아카이빙에서의 정보 개체를 효율적으로 기술하기 위해 정보자원 클러스터의 개념을 제안하고, 이를 실제적으로 구현하기 위해 XML 구문에 기반한 RDF를 적용하여 보존 기술항목 프레임워크를 구축하였다. 이는 OAIS 참조모델과 FRBR의 개념적인 구조를 결합하여 보존의 대상이 되는 정보자원을 보다 효과적으로 기술하고 보존 메타데이터를 생성할 수 있는 대안적인 접근방법으로 활용될 수 있을 것으로 기대된다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1994년도 Proceedings of the Korea Automatic Control Conference, 9th (KACC) ; Taejeon, Korea; 17-20 Oct. 1994
• /
• pp.731-736
• /
• 1994

Experiment on a lab-scale pH process is carried out to evaluate the control performance of the neural linearizing control scheme(NLCS) using a radial basis function(RBF) network which was previously proposed by Kim and Park. NLCS was developed to overcome the difficulties of the conventional neural controllers which occur when they are applied to chemical processes. Since NLCS is applicable for the processes which are already controlled by a linear controller and of which the past operating data are enough, we first control the pH process with PI controller. Using the operating data with PI controller, the linear reference model is determined by optimization. Then, a IMC controller replaces the PI controller as a feedback controller. NLCS consists of the IMC controller and a RBF network. After the learning of the neural network is fully achieved, the dynamics of the process combined with the neural network becomes linear and close to that of the linear reference model and the control performance of the linear control improves. During the training, NLCS maintains the stability and the control performance of the closed loop system. Experimental results show that the NLCS performs better than PI controller and IMC for both the servo and the regulator problems.

• 제어로봇시스템학회논문지
• /
• 제9권10호
• /
• pp.836-843
• /
• 2003

The automatic train control (ATC) system development project for the Automated Guideway Transit (AGT) system has high technical risk because the system is unmanned train control system using wireless technology which was unprecedented in train control industry of Korea. To overcome the technical risk during concept design phase of the ATC system development project, the integrated product team(IPT) carried out a reverse and reengineering process using a systems engineering design model. The generic systems engineering process is incorporated in the both reverse and reengineering process. As a result of the systems engineering effort, the IPT has built top layer systems engineering design model of the ATC subsystem. The purpose of this paper is to deliver the reverse and reengineering process which was used to develop the systems engineering design model of ATC system using a computer aided systems engineering tool. This study also shows that the model based reverse and reengineering process can reduce the technical risk by identifying the differences of requirement, functional and physical architecture between a reference system and a target system.

• 한국군사과학기술학회지
• /
• 제21권3호
• /
• pp.315-322
• /
• 2018

Due to the increasing demand for more rapid, precise and accurate geolocation of the targets on video frames from UAVs, an efficient and timely method for correcting sensor models of motion imagery is required. In this paper, we propose a method to adjust or correct sensor models of motion imagery frames using space resection via image matching with reference data. The proposed method adopts image matching between the motion imagery frames and the reference frames which are synthesized from reference data. Ground or reference control points are generated or selected through the matching process in near real time, and are used for space resection to get adjusted sensor models. Finally, more precise and accurate geolocation of the targets can possibly be done on the fly, and we have got the promising result on performance analysis in terms of the geolocation quality.

• 기술혁신학회지
• /
• 제8권spc1호
• /
• pp.323-336
• /
• 2005

Digital content industry is growing rapidly because of the property of high-speed networking and greater demand for digital fulltext-image. However, we know the fact that it is many difficulty in production and supply for good quality of content. Hereupon, we suggest digital fulltext-image service framework to protect copyright. More concretely, we propose integrated model and reference model to securely serve digital fulltext-image by recompositing core objects and reconstructing the value-chain structure of digital content industry to framework including the process(from its production (creators or copyrighters) to consumption (users or consumers)). Also, we construct the digital fulltext- image service system based on reference model and reconstruct its interface that occurs between core subjects.

• 대한전기학회논문지
• /
• 제36권10호
• /
• pp.748-754
• /
• 1987

A robustness analysis for model reference adaptive control(MRAC) system with plant uncertainty is discussed in this paper. The adaptive control system is designed under assumptions that the controlled plant is represented by a lst order nominal model and that the system is drived by a constant reference signal. When using general gradient method(GGM), it is shown that unmodelled dynamics in plant model can cause the instability of the overall control loop during the adaptation process. However, as the algorithm of least square method(LSM) is introduced, the global stability of the system can be hold. And it is also given that the boundedness of adjustable parameters may be verified using the concept of an equilibrium point analysis.

• Asia pacific journal of information systems
• /
• 제25권2호
• /
• pp.356-369
• /
• 2015

Public sectors around the world utilize social media tools and technologies in their daily activities for a variety of purposes, including disseminating useful information, fostering mass collaboration, and enforcing laws and regulations. A number of social media-based government stage models have emerged to document this use. In this chapter, we conducted a qualitative meta-synthesis of four social media-based government models. These models include 1) the open government maturity model, 2) the social media utilization model, 3) the adoption process for social media, and 4) the social media-based engagement model. The concepts, metaphors, and themes contained in these developmental models are extracted through a series of in-depth semantic analyses of descriptions, resulting in a common frame of reference.