• Proceedings of the KIEE Conference
• /
• 2004.07d
• /
• pp.2580-2582
• /
• 2004

In this paper, a Neural Network(NN) based approach for classification of two dimensions images. The proposed algorithm is able to apply in the actual industry. The described diagnostic algorithm is presented to defect surface failures on tiles. A way to get data for a digital image process is several kinds of it. The tiles are scanned and the digital images are preprocessed and classified using neural networks. It is important to reduce the amount of input data with problem specific preprocessing. The auto-associative neural network is used for feature generation and selection while the probabilistic neural network is used for classification. The proposed algorithm is evaluated experimentally using one hundred of the real tile images. Sample image data to preprocess have histogram. The histogram is used as input value of probabilistic neural network. Auto-associative neural network compress input data and compressed data is classified using probabilistic neural network. Classified sample images are determined by human state. So it is intervened human subjectivity. But digital image processing and neural network are better than human classification ability. Therefore it is very useful of quality control improvement.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.20 no.6
• /
• pp.183-194
• /
• 2010

A company's private network protected by a firewall and NAT(Network Address Translation) is not accessible directly through an external internet. However, as Reverse Connection technology used by NetCat extends to the technologies such as SSH Tunnel or HTTP Tunnel, now anyone can easily access a private network of corporation protected by a firewall and NAT. Furthermore, while these kinds of technologies are commercially stretching out to various services such as a remote control and HTTP Tunnel, security managers in a company or general users are confused under the circumstances of inner or outer regulation which is not allowed to access to an internal system with a remote control. What is more serious is to make a covert channel invading a company's private network through a malicious code and all that technologies. By the way, what matters is that a given security system such as a firewall cannot shield from these perceived dangers. So, we analyze the indirect access of technological methods and the status quo about a company's internal network and find a solution to get rid of the related dangers.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.01a
• /
• pp.15-18
• /
• 2014

하루가 다르게 최신의 기술이 쏟아져 나오는 정보화 시대에 들어서면서, 세계 ITC 분야는 디지털 기술과 유무선통신에 근간을 둔 인터넷의 확산으로 반도체, 컴퓨터, 콘텐츠 미디어, 정보가전 등 다양한 산업 분야가 융합되어 새로운 부가 가치를 창출하기에 이르렀다. 특히, 스마트 시대를 살아가는 우리의 일상생활은 모바일은 물론 일반적인 PC등을 사용한 인터넷 중심의 네트워크와 커뮤니케이션이 중요시 되고 있다. 이러한 시대의 흐름은 인프라 측면에서 네트워크의 트래픽을 폭증시키고, 많은 보안적인 문제를 야기하고 있다. 이에 임베디드 디바이스를 이용하여 네트워크를 운영하는 관리자에게 큰 도움을 주고, 업무 효율성을 높일 수 있는 포터블 형태의 네트워크 트래픽 탐지 시스템을 연구하게 되었다.

• The KIPS Transactions:PartA
• /
• v.15A no.5
• /
• pp.249-258
• /
• 2008

DCDN(Distributed Contents Delivery Network) as known the next generation of CDN is to exploit the resource of client PCs based on P2P technology to provide low cost, high scalability and high speed services when contents are delivered. This DCDN has two problems. One is that DCDN can't provide a sustainable service because client PCs are on/off irregularly and their on times tend to focus on specific time zone, and the other is that client PCs can not be accessed outside in DCDN if they are behind NAT or Firewall. In this paper, we present a distributed contents delivery network based on home gateway for a sustainable service. If home gateway is used in DCDN instead of client PC, it can solve the two existing problems of DCDN. That is, home gateway can provide a sustainable service because it is on during all day and it is not behind NAT and Firewall. The proposed method is implemented with ASUS WL-500GP, a wired/wireless router. Experimental results show the effectiveness of the proposed method compared to the existing method of DCDN.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2488-2491
• /
• 2000

A Neural network is possible to nonlinear function mapping and parallel processing. Therefore It has been developing for a Diagnosis system of nuclear plower plant. In general Neural Networks is a static mapping but Dynamic Neural Network(DNN) is dynamic mapping. When a fault occur in system, a state of system is changed with transient state. Because of a previous state signal is considered as a information. DNN is better suited for diagnosis systems than static neural network. But a DNN has many weights, so a real time implementation of diagnosis system is in need of a rapid network architecture. This paper presents a algorithm for RCP monitoring Alarm diagnosis system using Self Dynamic Neural Network(SDNN). SDNN has considerably fewer weights than a general DNN. Since there is no interlink among the hidden layer. The effectiveness of Alarm diagnosis system using the proposed algorithm is demonstrated by applying to RCP monitoring in Nuclear power plant.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.452-455
• /
• 2005

The cutting characteristics of hardened steel by a PCBN tool is investigated with respect to workpiece surface roughness, cutting force and tool flank wear of the vision system. Backpropagation neural networks (BPNs) were used for detection of tool wear. The neural network consisted of three layers: input, hidden and output. The input vectors comprised of spindle rotational speed, feed rates, vision flank wear, and thrust force signals. The output was the tool wear state which was either usable or failure. Hard turning experiments with various spindle rotational speed and feed rates were carried out. The learning process was performed effectively by utilizing backpropagation. The detection of the abnormal states using BPNs achieved 96.4% reliability even when the spindle rotational speed and feedrate were changed.

• Proceedings of the KOSOMBE Conference
• /
• v.1997 no.05
• /
• pp.430-433
• /
• 1997

We present the preliminary algorithms for automatic sleep scoring. According to the Rechtschaffen & Kales[3]'s critera, we developed six events detectors and eight parameters which contain the background information of signals, such as EEG, EMG, EOG. With the calculated parameters, we scored each epoch by IF-THEN rules, ANFIS for REM preiods, and finally Neural Network for unobvious epochs. The typical point of this algorithm is that the epoch which had good data sets were calculated in the first stage, and unobvious epochs were postponed until the final stage. After staging the good epochs, we classified unobvious epochs by the dominant stage of previous and posterior epochs.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.04b
• /
• pp.1073-1076
• /
• 2001

최근 홈 네트워크를 구축할 수 있는 기반 전송기술이 급속도로 발전하고 있다. 또한, 이를 이용할 수 있는 고성능의 가전기기가 속속 개발됨에 따라 흠 네트워크가 대중화되고 있다. 이러한 상황에서 가장 중요한 이슈로 떠오르고 있는 것이 홈 네트워크 보안이다. NAT기술은 내부 네트워크를 외부로 부터 은닉시키고 보호할 수 있는 새로운 기법을 제공한다는 점에서 최근 많은 관심을 받고 있다. 본 논문에서는 홈 네트워크 보안 시스템을 구현함에 있어 NAT기술과 범용 보안 메커니즘 IPSec을 적용하여 홈 네트워크를 인터넷과 같은 외부의 전산망에 존재하는 위험으로부터 안전하게 보호할 수 있는 기법을 제시한다.

• Proceedings of the KIEE Conference
• /
• 2001.04a
• /
• pp.84-86
• /
• 2001

This paper deals with the characteristic and thermal analysis of brushless motors for Rack assist type Electric Power Steering(REPS). The performance of permanent magnet is under the influence of temperature. To predict the motor performance, the thermal analysis is necessary. The equivalent thermal network is composed of the thermal resistance and the temperature of major parts is calculated according to the operating condition.

• Proceedings of the KIEE Conference
• /
• 2001.04a
• /
• pp.81-83
• /
• 2001

This paper introduces the improved analysis method for PMDC (Permanent Magnet DC) motors. The effect of field yoke is considered by equivalent area. Overhang effect, caused by different axial length of magnet and armature, is analyzed by 3D EMCN(Equivalent Magnet Circuit Network) method. Then brush contact resistance and mechanical loss extracted by experimental data are used in characteristic analysis. The validity of the proposed method is verified by the comparison of analysis results and experimental ones.