• The Journal of the Korea Contents Association
• /
• v.9 no.10
• /
• pp.110-121
• /
• 2009

In this paper, we propose the new method that offer detailed-information through relax the system memory limitation about 3D model to user. That method based on making LOD(Level of Detail) model from huge 3D data of structure cultural assets. In our method as transformed AOSP algorithm, first of all it create the hierarchical structure space about 3D data, and create the LOD model by surface simplification. Then it extract the ROI(Region of Interest) of user in simplified LOD model, and then do rendering by original model and same surface detailed-information after process the local detailed in extracted region. To evaluate the proposed method, we have some experiment by using the precise 3D scan data of structure cultural assets. Our method can offer the detailed-information same as exist method, and moreover 45% reduced consumption of memory experimentally by forming mesh structure same as ROI of simplified LOD model. So we can check the huge structure cultural assets particularly in general computer environment.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.7 no.5
• /
• pp.97-105
• /
• 2008

The micro capacitors are passive elements necessary to electronic circuits and wireless portable PAN(personal area network) and Mobile Communications device modules in the baseband circuits in combination with another passive and active devices. As capacitance is proportionally increased with dielectric constant and electrode areas, in addition, inversely decreased the thickness of the dielectric material, thus thin film capacitors are generally seen as a preferable means to achieve high performance and thin film capacitors are used in a variety of functional circuit devices. In this paper, propose dielectric material as AIN(Aluminium nitride) to make micro thin film capacitor, and this capacitor has the MIM(metal-insulator-metal) structure. AIN thin films are widespread applied because they had more excellent properties such as chemical stability, high thermal conductivity, electrical isolation and so on. In addition, AIN films show low frequency response for baseband signal ranges, I-V and C-V electrical characterization of a thin film micro capacitor. The above experimental test and estimated results demonstrate that the thin film capacitor has sufficient and efficient functional performance to be the baseband range frequency of general electronics circuit and passive device applications.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.5
• /
• pp.1155-1166
• /
• 2015

It is almost impossible to store or analyze big data increasing exponentially with traditional technologies, so Hadoop is a new technology to make that possible. In recent R is using as an engine for big data analysis based on distributed processing with Hadoop technology. With RHadoop that integrates R and Hadoop environment, we implemented parallel multiple regression analysis with various data sizes of actual data and simulated data. Experimental results showed our RHadoop system was faster as the number of data nodes increases. We also compared the performance of our RHadoop with lm function and biglm packages available on bigmemory. The results showed that our RHadoop was faster than other packages owing to paralleling processing with increasing the number of map tasks as the size of data increases.

• Journal of the Korea Society for Simulation
• /
• v.19 no.3
• /
• pp.7-16
• /
• 2010

In this paper, we propose an ambient light control method based on the difference of image frames in video. The proposed method is composed of three steps. 1) The first step is to extract a dominant color of a current frame. 2) The second step is to compute the amount of change and the representative color in the changed region using the difference image. 3) The third step is to make a new representative color. The difference image is created from two images transformed into the YUV color space. The summed color difference of each pixel is used for the amount of change. The new representative color is created by synthesizing the current color and the changed color in proportion to the amount of change. We compare the variations of the light effect according to time with and without the proposed method for the same video. The result shows that the new method generates more dynamic light effects.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.4
• /
• pp.213-220
• /
• 2024

This paper proposes an optimization strategy for performing Federated Learning between devices and edge clouds using WebAssembly (WASM). The proposed strategy aims to maximize efficiency by conducting partial training on devices and the remaining training on edge clouds. Specifically, it mathematically describes and evaluates methods to optimize data transfer between GPU memory segments and the overlapping of computational tasks to reduce overall training time and improve GPU utilization. Through various experimental scenarios, we confirmed that asynchronous data transfer and task overlap significantly reduce training time, enhance GPU utilization, and improve model accuracy. In scenarios where all optimization techniques were applied, training time was reduced by 47%, GPU utilization improved to 91.2%, and model accuracy increased to 89.5%. These results demonstrate that asynchronous data transfer and task overlap effectively reduce GPU idle time and alleviate bottlenecks. This study is expected to contribute to the performance optimization of Federated Learning systems in the future.

• Journal of KIISE
• /
• v.42 no.10
• /
• pp.1195-1206
• /
• 2015

For the management of a knowledge system, systems that automatically infer and manage scalable knowledge are required. Most of these systems use ontologies in order to exchange knowledge between machines and infer new knowledge. Therefore, approaches are needed that infer new knowledge for scalable ontology. In this paper, we propose an approach to perform rule based reasoning for scalable SHIF ontologies in a spark framework which works similarly to MapReduce in distributed memories on a cluster. For performing efficient reasoning in distributed memories, we focus on three areas. First, we define a data structure for splitting scalable ontology triples into small sets according to each reasoning rule and loading these triple sets in distributed memories. Second, a rule execution order and iteration conditions based on dependencies and correlations among the SHIF rules are defined. Finally, we explain the operations that are adapted to execute the rules, and these operations are based on reasoning algorithms. In order to evaluate the suggested methods in this paper, we perform an experiment with WebPie, which is a representative ontology reasoner based on a cluster using the LUBM set, which is formal data used to evaluate ontology inference and search speed. Consequently, the proposed approach shows that the throughput is improved by 28,400% (157k/sec) from WebPie(553/sec) with LUBM.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.310-318
• /
• 2018

Recently, the importance of big data analysis is increasing as a large amount of data is generated by various devices connected to the Internet with the advent of Internet of Things (IoT). Especially, it is necessary to analyze various large-scale IoT streaming sensor data generated in real time and provide various services through new meaningful prediction. This paper proposes a real-time indoor PM10 concentration prediction LSTM model based on streaming data generated from IoT sensor using AWS. We also construct a real-time indoor PM10 concentration prediction service based on the proposed model. Data used in the paper is streaming data collected from the PM10 IoT sensor for 24 hours. This time series data is converted into sequence data consisting of 30 consecutive values from time series data for use as input data of LSTM. The LSTM model is learned through a sliding window process of moving to the immediately adjacent dataset. In order to improve the performance of the model, incremental learning method is applied to the streaming data collected every 24 hours. The linear regression and recurrent neural networks (RNN) models are compared to evaluate the performance of LSTM model. Experimental results show that the proposed LSTM prediction model has 700% improvement over linear regression and 140% improvement over RNN model for its performance level.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.25 no.1
• /
• pp.83-91
• /
• 2014

The passive UHF RFID tags for item-level tagging are now explosively used in the supply chain and retail applications as they have several advantages, the two most relevant are cost and a longer read range. However, the signal interference problem between RFID tags and smart devices in real world is expected according to the smart-phone and tablet market growth together. The performance of RFID tags can be significantly less. The popular examples are the read-success rates and read range reduction. Especially, KT Corp. recently emphasized the serious signal interference at 900 MHz of LTE and old RFID frequencies through their public demonstration. By popular demands, this paper suggests the interference tolerance measurement method between the passive UHF RFID tag and the transmitted signal from a smart device. In addition, we selected three passive UHF RFID tags(Inlay) available on the market and quantitatively evaluated read range reduction results by interference signals using the PCR(Performance Change Rates) index. As a result, the LTE system is about three times as effective as the WCDMA system in terms of interference effects, and the read range performance of two RFID tags about 60 % drop.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.29 no.7
• /
• pp.902-909
• /
• 2023

This study evaluated the performance test of a small-sized LTE-Maritime(LTE-M) transceiver that was developed and promoted to expand the use of intelligent maritime traf ic information services led by the Ministry of Oceans and Fisheries with the aim of supporting the prevention of maritime accidents. Accoriding to statistics, approximately 30% of all marine accidents in Korean water occur with ships weighing less than 3 tons. Therefore, the blind spots of maritime safety must be supplemented through the development of small-sized transceivers. The small transceiver may be used in fishing boats that are active near coastal waters and in water leisure equipment near the coastline. Therefore, verifying whether sufficient performance and stable communication quality are provided is necessary, considering the environment of their real usage. In this study, we reviewed the communication quality goals of the LTE-M network and the performance requirements of small-sized transceivers suggested by the Ministry of Oceans and Fisheries, and proposed a test plan to appropriately evaluate the performance of small-sized transceivers. The validity of the proposed test method was verified for six real-sea areas with a high frequency of marine accidents. Consequently, the downlink and uplink transmission speeds of the small-sized LTE-M transceiver showed performances of 9 Mbps or more and 3 Mbps or more, respectively. In addition, using the coverage analysis system, coverage of more than 95% and 100% were confirmed in the intensive management zone (0-30 km) and interesting zone (30-50 km), respectively. The performance evaluation method and test results proposed in this paper are expected to be used as reference materials for verifying the performance of transceivers, contributing to the spread of government-promoted e-navigation services and small-sized transceivers.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10c
• /
• pp.253-257
• /
• 2007

최근 웹 상에 정보가 폭발적으로 증가함에 따라, 사용자의 취향에 맞는 정보를 선별하여 제공하는 추천 시스템에 대한 연구가 활발히 진행되고 있다. 추천시스템은 사용자의 관심정보를 기술한 사용자 프로파일을 기반으로 동작하기 때문에 정확한 사용자 프로파일의 생성은 매우 중요하다. 사용자의 암시적인 행동정보를 기반으로 취향을 분석하는 대표적인 연구로 사용자가 이용한 웹 문서를 분석하는 방법이 있다. 이는 사용자가 이용하는 웹 문서에 빈번하게 등장하는 단어를 기반으로 사용자의 프로파일을 생성하는 것이다. 그러나 최근 웹 문서는 사용자 취향과 관련 없는 많은 구성요소들(로고, 저작권정보 등)을 포함하고 있다. 따라서 이러한 내용들을 모두 포함하여 웹 문서를 분석한다면 생성되는 프로파일의 정확도는 낮아질 것이다. 따라서 본 논문에서는 사용자 기기에서 사용자의 웹 문서 이용내역을 분석하고, 동일한 사이트로부터 얻어진 문서들에서 반복적으로 등장하는 블록을 제거한 후, 정보블럭을 식별하여 사용자의 관심단어를 추출하는 새로운 프로파일 생성방법을 제안한다. 이를 통해 보다 정확하고 빠른 프로파일 생성이 가능해진다. 본 논문에서는 제안방법의 평가를 위해, 최근 구매활동이 있었던 사용자들이 이용한 웹 문서 데이터를 수집하였으며, TF-IDF 방법과 제안방법을 이용하여 사용자 프로파일을 각각 추출하였다. 그리고 생성된 사용자 프로파일과 구매데이터와의 연관성을 비교하였으며, 보다 정확한 프로파일이 추출되는 결과와 프로파일 분석시간이 단축되는 결과를 통해 제안방법의 유효성을 입증하였다.)으로 높은 점수를 보였으며 내장첨가량에 따른 관능특성에서는 온쌀죽은 내장 $2{\sim}5%$ 첨가, 반쌀죽은 내장 $3{\sim}5%$ 첨가구에서 유의적(p<0.05)으로 높은 점수를 보였으나 쌀가루죽은 내장 $1{\sim}2%$ 첨가구에서 유의적(p<0.05)으로 낮은 점수를 보였다. 이상의 연구 결과를 통해 온쌀은 2%, 반쌀은 3%, 쌀가루는 4%의 내장을 첨가하여 제조한 전복죽이 이화학적, 물성적 및 관능적으로 우수한 것으로 나타났다.n)방법의 결과와 비교하였다.다. 유비스크립트에서는 모바일 코드의 개념을 통해서 앞서 언급한 유비쿼터스 컴퓨팅 환경에서의 문제점을 해결하고자 하였다. 모바일 코드에서는 프로그램 코드가 네트워크를 통해서 컴퓨터를 이동하면서 수행되는 개념인데, 이는 물리적으로 떨어져있으면서 네트워크로 연결되어 있는 다양한 컴퓨팅 장치가 서로 연동하기 위한 모델에 가장 적합하다. 이는 기본적으로 배포(deploy)라는 단계가 필요 없게 되고, 새로운 버전의 프로그램이 작성될지라도 런타임에 코드가 직접 이동하게 되므로 버전 관리의 문제도 해결된다. 게다가 원격 함수를 매번 호출하지 않고 한번 이동된 코드가 원격지에서 모두 수행을 하게 되므로 성능향상에도 도움이 된다. 장소 객체(Place Object)와 원격 스코프(Remote Scope)는 앞서 설명한 특징을 직접적으로 지원하는 언어 요소이다. 장소 객체는 모바일 코드가 이동해서 수행될 계산 환경(computational environment