• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.97-100
• /
• 2016

Using the CNN(Convolutinal Neural Network), Deep Learning for variety of fields are being developed and these are showing significantly high level of performance at image recognition field. In this paper, we show the test accuracy which is learned by large-scale training data, over 5,000,000 of Korean characters. The architecture of CNN used in this paper is KCR(Korean Character Recognition)-AlexNet newly created based on AlexNet. KCR-AlexNet finally showed over 98% of test accuracy. The experimental data used in this paper is large-scale Korean character database PHD08 which has 2,187 samples for each Korean character and there are 2,350 Korean characters that makes total 5,139,450 sample data. Through this study, we show the excellence of architecture of KCR-AlexNet for learning PHD08.

• The KIPS Transactions:PartA
• /
• v.14A no.4
• /
• pp.215-226
• /
• 2007

For past years, many distributed virtual shared-memory(DVSM) systems have been studied in order to develop a low-cost shared memory system with a fast interconnection network. But the DVSM needs a lot of data and control communication between distributed processing nodes in order to provide memory consistency in software, and this communication overhead significantly dominates the overall performance. In general, the communication overhead also increases as the number of processing nodes increase, so communication overhead is a very important performance factor for developing a large-scale DVSM system. In this paper, we study the performance scalability quantitatively and qualitatively for developing a large-scale DVSM system based on the next generation interconnection network, called the InfiniBand. Based on the study, we analyze a performance requirement of the next-coming interconnection network to be used for developing a performance-scalable DVSM system in the future.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.1
• /
• pp.114-121
• /
• 2022

In this paper, we consider a user centric clustering in order to guarantee the performance of the users in cell free multiple-input multiple-output (MIMO) network. In the user centric clustering scheme, by using large scale fading coefficients of the connected access points (APs), each user decides own cluster with the APs having the higher the large scale fading coefficients than threshold value compared to the highest large scale fading coefficient. In the determined user centric clusters, the APs design the beamformers and power allocations in the distributed manner and the APs cooperatively transmit data to users by using beamformers and power allocations. In the simulation results, we verify the performance of user centric clustering in terms of the spectral efficiency and we also find the optimal threshold value in the given configuration.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.12
• /
• pp.1113-1119
• /
• 2000

To operate a process plant safely and economically, process monitoring is very important. Process monitoring is the task to identify the state of the system from sensor data. Process monitoring includes data acquisition, regulatory control, data reconciliation, fault detection, etc. This research focuses on the data recon-ciliation using scale-space filtering and fault detection using functional-link associative neural networks. Scale-space filtering is a multi-resolution signal analysis method. Scale-space filtering can extract highest frequency factors(noise) effectively. But scale-space filtering has too large calculation costs and end effect problems. This research reduces the calculation cost of scale-space filtering by applying the minimum limit to the gaussian kernel. And the end-effect that occurs at the end of the signal of the scale-space filtering is overcome by using extrapolation related with the clustering change detection method. Nonlinear principal component analysis methods using neural network have been reviewed and the separately expanded functional-link associative neural network is proposed for chemical process monitoring. The separately expanded functional-link associative neural network has better learning capabilities, generalization abilities and short learning time than the exiting-neural networks. Separately expanded functional-link associative neural network can express a statistical model similar to real process by expanding the input data separately. Combining the proposed methods-modified scale-space filtering and fault detection method using the separately expanded functional-link associative neural network-a process monitoring system is proposed in this research. the usefulness of the proposed method is proven by its application a boiler water supply unit.

• Smart Structures and Systems
• /
• v.6 no.5_6
• /
• pp.525-543
• /
• 2010

Testing and validation processes are critical tasks in developing a new hardware platform based on a new technology. This paper describes a series of experiments to evaluate the performance of a newly developed MEMS-based wireless sensor node as part of a wireless sensor network (WSN). The sensor node consists of a sensor board with four accelerometers, a thermometer and filtering and digitization units, and a MICAz mote for control, local computation and communication. The experiments include calibration and linearity tests for all sensor channels on the sensor boards, dynamic range tests to evaluate their performance when subjected to varying excitation, noise characteristic tests to quantify the noise floor of the sensor board, and temperature tests to study the behavior of the sensors under changing temperature profiles. The paper also describes a large-scale deployment of the WSN on a long-span suspension bridge, which lasted over three months and continuously collected ambient vibration and temperature data on the bridge. Statistical modal properties of a bridge tower are presented and compared with similar estimates from a previous deployment of sensors on the bridge and finite element models.

• Communications for Statistical Applications and Methods
• /
• v.29 no.2
• /
• pp.161-176
• /
• 2022

There has been active research in image classification using deep learning convolutional neural network (CNN) models. ImageNet large-scale visual recognition challenge (ILSVRC) (2010-2017) was one of the most important competitions that boosted the development of efficient deep learning algorithms. This paper introduces and compares six monumental models that achieved high prediction accuracy in ILSVRC. First, we provide a review of the models to illustrate their unique structure and characteristics of the models. We then compare those models under a unified framework. For this reason, additional devices that are not crucial to the structure are excluded. Four popular data sets with different characteristics are then considered to measure the prediction accuracy. By investigating the characteristics of the data sets and the models being compared, we provide some insight into the architectural features of the models.

• Journal of the Korea Society for Simulation
• /
• v.15 no.1
• /
• pp.87-95
• /
• 2006

We need to check into when a security system is newly developed, we against cyber attack which is expected in real network. However it is impossible to check it under the environment of a large-scale distributive network. So it is need to simulate it under the virtual network environment. SSFNet is a event-driven simulator which can be represent a large-scale network. Unfortunately, it doesn't have the module to simulate security functions. In this paper, we added the IDS module to SSFNet. We implement the IDS module by modeling a key functions of Snort. In addition, we developed some useful functions using Java language which can manipulate easily a packet for network simulation. Finally, we performed the simulation to verify the function if our developed IDS and Packets Manipulation. The simulation shows that our expanded SSFNet can be used to further large-scale security system simulator.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.1
• /
• pp.80-90
• /
• 1996

The efficienty of ATM is based on the statical multiplexing of fixed-length packets, which are called cells. The most important technical point for realizing ATM switching network is an arrangement of the buffers and switches. Current most ATM switching networks are being achieved by using the switching modules based on the unit switch of $8{\times}8$ 150Mb/s or $16{\times}16$ 150Mb/s, the unit switch of $32{\times}32$150Mb/s for a large scale system is under study in many countries. In this paper, we proposed a new $32{\times}32$(4.9Gb/s throughput) ATM switch using Shared buffer memory switch which provides superior traffic characteristics in the cell loss, delay and throughput performance and easy LSI(Large Scale Integrated circuit). We analytically estimated and simulated by computer the buffer size into it. We also proposed the configuration of the large capacity ATM switching network($M{\times}M$.M>1,000) consisting of multistage to improve the link speed by non-blocking.

• Korean Management Science Review
• /
• v.17 no.1
• /
• pp.55-64
• /
• 2000

By managing not only geographical information but also various kinds of attribute data. GIS presents useful information for decision-makings. Most of decision-making problems using GIS can be formulated into network-optimization problems. In this study we deal with the implementation of network optimization system that extracts data from the database in GIS. solves a network optimization problem and present optimal solutions through GIS' graphical user interface. We design a nitwork optimization system and present some implementation techniques by showing a prototype of the network optimization system. Our network optimization system consists of three components : the interface module for user and GIS the basic network the program module the advanced network optimization program module. To handle large-scale networks the program module including various techniques for large sparse networks is also considered, For the implementation of the network optimization system we consider two approaches : the method using script languages supported by GIS and the method using client tools of GIS. Finally some execution results displayed by the prototype version of network optimization system are given.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2002.06a
• /
• pp.97-106
• /
• 2002

Grid makes it possible to cooperate with other network area by sharing and using distributed resources. In order to manage effectively large-scale Grid network resources, Grid NOC needs monitoring architecture that can manage distributed resources in one time. Being restricted within specific managing area, conventional network management system has limitation in extension of managing area and in general management of heterogeneous resource. In this paper, we design a monitoring architecture that can take in the situation and has scalability. In the monitoring architecture the network areas publish information in a common directory service, and then Grid NOC can connect to the network areas directly by using this information. Therefore, it makes us possible to manage overall large-scale resource of Grid network reducing load.