• The KIPS Transactions:PartC
• /
• v.9C no.3
• /
• pp.337-342
• /
• 2002

In this paper, we proposed the ISL for evaluating the security level of Intranet. This method is composed of two parts First, Hacking Scenarios are the hacking stories by hacking/hackers levels. Second, Network Model is composed of the zone and intrusion paths. Our method is very active and practical. We divide ISL into five levels ISL1∼ISL5. Among the levels, ISL5 is the highest security level. Main purpose of this paper is to promote the security of Intranet. In this paper, we designed and analyzed the test-bed for ISL3.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.4
• /
• pp.133-140
• /
• 2008

The cable TV network has delivered downward broadcasting signals from distribution centers to subscribers. Since the traditional coaxial cable has been upgraded by the Hybrid Fiber Coaxial(HFC) cable, the upward channels has expanded broadband services such as Internet. This upward channel is vulnerable to ingress noises. When the noises from the children nodes accumulated in an amplifier exceeds a certain level, that node has to be cut off to prevent the noise propagation. The node selection problem(NSP) is defined to select nodes so that the noise in each node does not exceed the given threshold value and the sum of Profits of selected nodes can be maximized. The NSP has shown to be NP-hard. In this paper, we have proposed heuristics to find the near-optimal solution for NSP. The experimental results show that interval partitioning is better than greedy approach. Our heuristics can be used by the HFC network management system to provide privileged services to the premium subscribers on HFC networks.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.3
• /
• pp.210-218
• /
• 2002

This paper proposes an efficient separation method of the composite images by using independent component analysis(ICA) based on neural networks of the approximate learning algorithm. The Proposed learning algorithm is the fixed point(FP) algorithm based on Secant method which can be approximately computed by only the values of function for estimating the root of objective function for optimizing entropy. The secant method is an alternative of the Newton method which is essential to differentiate the function for estimating the root. It can achieve a superior property of the FP algorithm for ICA due to simplify the composite computation of differential process. The proposed algorithm has been applied to the composite signals and image generated by random mixing matrix in the 4 signal of 500-sample and the 10 images of $512{\times}512-pixel$, respectively The simulation results show that the proposed algorithm has better performance of the learning speed and the separation than those using the conventional algorithm based method. It also solved the training performances depending on initial points setting and the nonrealistic learning time for separating the large size image by using the conventional algorithm.

• The KIPS Transactions:PartC
• /
• v.12C no.3 s.99
• /
• pp.449-456
• /
• 2005

Optical burst switching (OBS) overcomes the inefficient resource usage of optical circuit switching and minimizes the optical buffering requirement of optical packet switching. General switch management protocol (GSMP) is an open interface between a label switch and a controller, and it provides connection, configuration, performance, event management and synchronization. GSMP open interface in the OBS network allows the implementation of OBS switch to be simple by separating the data forward plane from the control plane. We defined managed objects to support connection, configuration, performance, and fault management for the management of OBS network in the GSMP open interface. We proposed the network management model, in which the above managed objects are distributed in a controller and an OBS switch according to network management functions. We verified the possibility of connection management using distributed network management model in the GSMP open interface of OBS network by implementing GSMP and network management functions with managed objects of OBS.

• Journal of KIISE:Software and Applications
• /
• v.33 no.12
• /
• pp.999-1007
• /
• 2006

The Backpropagation learning algorithm is known to have slow and false convergence aroused from plateau and local minima. Many substitutes for backpropagation announced so far appear to pay some trade-off for convergence speed and stability of convergence according to parameters. Here, a new algorithm is proposed, which avoids some of those problems associated with the conventional backpropagation problems, especially with local minima, and gives relatively stable and fast convergence with low storage requirement. This is the separate learning algorithm in which the upper connections, hidden-to-output, and the lower connections, input-to-hidden, separately trained. This algorithm requires less computational work than the conventional backpropagation and other improved algorithms. It is shown in various classification problems to be relatively reliable on the overall performance.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.1
• /
• pp.167-174
• /
• 1999

In this paper, we propose the adaptive separated-queueing ATM switching model for the effective processing of the various types of multimedia traffic by virtue of ATM switching network with multiple outlets which is the essential part of B-ISDN. This proposed model employs the dynamically separated buffering mechanism in the processing of two classes of cell, realtime service traffic and non-realtime service traffic, at the output buffer to enhance the overall QoS(Quality of Service). The adopted ATM switch architecture has Batcher-banyan based network, but it uses different topologies and control techniques 6to resolve the cell contention. For the performance evaluation of our proposed method, we have done both analytical modeling and simulation. The both results show that our proposed queueing strategy is quite appropriate to the ATM switch with multiple outlets and can quarantee the QoS requirements of the incoming multimedia traffic.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.15 no.7
• /
• pp.618-627
• /
• 1990

In this paper a new algorithm for separation of contacting subpattern and connective feature extraction of strokes is proposed. This algorithm is able to classification of the type of contacting parts, connective feature extreaction of strokes, separate the phoneme of contacting parts between strokes, classify the character types by feature classification of connecting parts and analysis of connecting attribute. Also, shape normalize into formal patterns and decide on the input pattern from position value of bending feature of this normalized shape and make an recognition experiment by neural network using BEP learining algorithm. This algorithm represents the good achievement ratio by separation of phoneme, classification of character type, connective feature extraction of stroke and recognition experiment.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.25-30
• /
• 2018

기존의 검색 기반 챗봇 시스템과 다르게 생성 기반 챗봇 시스템은 사전에 정의된 응답에 의존하지 않고 채팅 말뭉치를 학습한 신경망 모델을 사용하여 응답을 생성한다. 생성 기반 챗봇 시스템이 사람과 같이 자연스러운 응답을 생성하려면 이전 문맥을 반영해야 할 필요가 있다. 기존 연구에서는 문맥을 반영하기 위해 이전 문맥과 입력 발화를 통합하여 하나의 벡터로 표현했다. 이러한 경우 이전 문맥과 입력 발화가 분리되어 있지 않아 이전 문맥이 필요하지 않는 경우 잡음으로 작용할 수 있다. 본 논문은 이러한 문제를 해결하기 위해 입력 발화와 이전 문맥을 각각의 벡터로 표현하는 방법을 제안한다. 또한 생성적 적대적 신경망을 통해 챗봇 시스템을 보강하는 방법을 제안한다. 채팅 말뭉치(55,000 개의 학습 데이터, 5,000개의 검증 데이터, 5,260 개의 평가 데이터)를 사용한 실험에서 제안한 문맥 반영 방법과 생성적 적대적 신경망을 통한 챗봇 시스템 보강 방법은 BLEU와 임베딩 기반 평가의 성능 향상에 도움을 주었다.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.40 no.12
• /
• pp.94-105
• /
• 2003

OBS(Optical Burst Switching) networks which decouple the burst from its header has characteristic one-way reservation. Therefore, the fault of a fiber link may lead to the magnitude of burst loss. Since sending a header packet and using an offset time before the transmission of the corresponding burst. In this paper, we analysis a offset time based on the burst loss rate resulting from only link failure and by applying the restoration schemes in an optical burst switching networks. The simulation results in terms of the effect of a offset time is evaluated in our network model to verify that comparing the numerical analysis with simulation result. Finally, and through these results, We propose the new restoration scheme based on offset time considered Quality of Service.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.7A
• /
• pp.1064-1073
• /
• 1999

In the integrated systems of wired and wireless networks, it is not fit to separate cellular networks to intelligent networks so IMT-2000 is basically developed including the intelligent networks. But it is not yet determined how functional entities of IMT-2000 map the corresponding physical entities. In this paper, we propose the integrated model by mapping various functional entities of IMT-2000 into the corresponding physical entities. Also, we proposs the call model for the mobile call setup and IN service call. Based on the proposed model, we evaluate the performance and compare with IMT-2000 in terms of the mean service delay and processing delay at each physical entities.