• 디지털산업정보학회논문지
• /
• 제15권4호
• /
• pp.31-40
• /
• 2019

This paper proposes an adaptive PCIe system for TI C66x DSPs. Conventionally, the PCIe system provided by the C66x is a system dependent on the structure in which the primary core writes an application to the DSP memory through the PCIe interface, then activate the secondary core. Due to the dependency between the cores, when developing a project using a PCIe interface, the remaining cores have to be programmed with a concern of the primary core used as the PCIe interface. Therefore, in order to de-couple the connections among the cores, an adaptive PCIe system is proposed, in the paper, in which the cores operate independently compared to the conventional system. Since the core used as the PCIe interface only runs PCIe related operations in the new system, the remaining cores can be fully utilized without concerning the connections with the core for PCIe interface. In order to verify the feasibility of the proposed adaptive PCIe system, the implementations of LTE-A down link, and IEEE 802.11ac are carried out using the evaluation board which includes a TMS320C6670 chip. Altogether, these results support that we demonstrated that the digital signal processing systems with the PCIe Interface can be developed more rapidly by applying the proposed system.

• 정보처리학회논문지A
• /
• 제8A권4호
• /
• pp.299-310
• /
• 2001

본 연구에서는 리눅스 가상 서버(LVS：Linux Virtual Server) 클러스터 시스템에서 실제 서버의 상태에 기초한 사용자의 요청을 분배하는 방법론을 연구하고자 한다. LVS 클러스터 시스템에서 사용자의 요청을 분배하는데 적용되는 기존 WLC(Weighted Least Connection) 방법론이 검토되었고, 실제 서버의 부하를 고려하여 각 서버에 요청을 할당하는 부하 분산 방법론을 제안하고자 한다. 부하 측정을 위한 실험은 가상의 부하를 생성하는 툴을 사용하여 서버에 부하를 부과하여 실행되었다. 본 연구에서 제시된 부하 분산 방법론이 기존의 WLC 방법론보다 실제 서버의 메모리 사용측면에서 효율을 기대할 수 있어 제안하고자 하며, 또한 서버 자원을 균형적으로 분배시키고 가중치의 변화에 대한 교정력(correction potentiality)이 어느 정도 개선됨을 확인할 수 있었다.

• 펄프종이기술
• /
• 제40권1호
• /
• pp.55-61
• /
• 2008

The annals of Joseon Dynasty is the most valuable resource containing the record of Korean history and has been known as one of the UNESCO's Memory of the World Register. In some period, wax-treated paper was used in the annals of Joseon Dynasty to prevent the damage from pests. But, wax-treated volumes were discolored to brown, white or black, and partially cracked by the stiffened wax. It means that wax-treated volumes were damaged by the hardening and oxidizing of wax. The wax is considered as the first reason in damages on the wax-treated volume. Therefore, in this study we analysed the oxidation index of Korea traditional paper(Hanji) which is used as the restoration paper of the waxed volume in the annals of Joseon Dynasty by using FT-IR. As the analysis of FT-IR, the traditional two-layed stamping paper (Hanji sample F), which was treated by starch in its surface, was shown the most stable state after the accelerated ageing test.

• 한국산업융합학회 논문집
• /
• 제24권3호
• /
• pp.323-331
• /
• 2021

With global warming and pollution problems, accurate forecasting of the harmful gases would be an essential alarm in our life. In this paper, we forecast the emission of the five gases(SOx, NO₂, NH₃, H₂S, CH₄) using the time series model of ARIMA, the learning algorithms of Random forest, and LSTM. We find that the gas emission data depends on the short-term memory and behaves like a random walk. As a result, we compare the RMSE, MAE, and MAPE as the measure of the prediction performance under the same conditions given to three models. We find that ARIMA forecasts the gas emissions more precisely than the other two learning-based methods. Besides, the ARIMA model is more suitable for the real-time forecasts of gas emissions because it is faster for modeling than the two learning algorithms.

• 전기전자학회논문지
• /
• 제26권4호
• /
• pp.545-552
• /
• 2022

특정 분야의 특허출원수는 기술의 수명주기 및 산업의 활성화 정도와 밀접한 관계를 가지고 있다. 따라서 사전에 사업을 준비하는 기업들과 미래 유망 기술을 초기 단계에서 선발하여 투자하고자 하는 정부 기관들은 미래의 특허 출원수 예측에 대해 큰 관심을 가지고 있다. 본 논문에서는 시계열 데이터에 적합한 RNN의 기법 중 하나인 양방향 LSTM 기법을 이용하여 기존 예측 방법들보다 정확도를 높이는 방법을 제안한다. 5개 분야의 대한민국 특허 출원 데이터에 대해서 제안된 방법은 기존에 사용되던 확산 모델 중 하나인 Bass 모델과 비교하여 평균 절대 백분율 오차(MAPE)의 값이 약 16퍼센트 향상된 결과를 보여준다.

• 디지털산업정보학회논문지
• /
• 제19권4호
• /
• pp.97-115
• /
• 2023

In South Korea, chronic kidney disease(CKD) impacts around 4.6 million adults, leading to a high reliance on hemodialysis. For effective dialysis, vascular access is crucial, with decisions about vascular surgeries often made during dialysis sessions. Anticipating these needs could improve dialysis quality and patient comfort. This study investigates the use of Artificial Intelligence(AI) to predict the timing of surgeries for dialysis vessels, an area not extensively researched. We've developed an AI algorithm using predictive maintenance methods, transitioning from machine learning to a more advanced deep learning approach with Long Short-Term Memory(LSTM) models. The algorithm processes variables such as venous pressure, blood flow, and patient age, demonstrating high effectiveness with metrics exceeding 0.91. By shortening the data collection intervals, a more refined model can be obtained. Implementing this AI in clinical practice could notably enhance patient experience and the quality of medical services in dialysis, marking a significant advancement in the treatment of CKD.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제12권2호
• /
• pp.219-229
• /
• 2012

In the course of feasibility study the necessity of implementing electrochemical methods as an inline metrology technique to characterize semiconductor nano structures for a Deep Trench Dynamic Random Access Memory (DT-DRAM) (e.g. ultra shallow junctions USJ) was discussed. Hereby, the state of the art semiconductor technology on the advantages and disadvantages of the most recently used analytical techniques for characterization of nano electronic devices are mentioned. Various electrochemical methods, their measure relationship and correlations to physical quantities are explained. The most important issue of this paper is to prove the novel usefulness of the electrochemical micro cell in the semiconductor industry.

• 경영과학
• /
• 제18권1호
• /
• pp.1-13
• /
• 2001

Despite that the semiconductor industry plays an important role to our economy, it has abnormal industrial structure stressing too much on memory chips. Thus, it is essential for our corporate to develop nonmemory chips to obtain technological leadership in a highly competitive semiconductor market. In this study, we demonstrate how benefit/cost analysis using the Analytic Hierarchy process (AHP) can be used for the proper selection of nonmemory semiconductor chips: Microprocessor, ASIC, digital IC and Analogue IC. The final results show that ASIC is the most attractive chip to develop, followed by Analogue IC, digital IC and Microprocessor. This is Somewhat consistent with the information that we found with respect to the elements that were taken into consideration. Sensitivity analysis is also provided here.

• 지식경영연구
• /
• 제12권2호
• /
• pp.81-90
• /
• 2011

This paper aims at the application of knowledge management in teaching Korean traditional dance (KTD) for elderly people. KTD has been taught using the traditional method of individual verbal instructions. Applying the traditional method in teaching a group of elderly people has limitations for both instructors and for the elderly people. Because there is no uniform method in teaching KTD, each instructor has to create his or her own teaching plan. Also, due to diminishing memory among the elderly people, it is difficult for them to recall what they were verbally instructed. To overcome these limitations in teaching KTD, we propose the application of knowledge management (KM) process. The KM process will allow collecting, storing, sharing, and optimizing the tacit knowledge of KTD. The findings and implications in this study would provide insights not only for the KTD instructors but for the silver industry as well.

• Asia pacific journal of information systems
• /
• 제2권1호
• /
• pp.57-76
• /
• 1992

Material Requirement Planning(MRP) has been the most widely implemented large scale production management system in the manufacturing industry. Computerization of MRP systems involves, in general, Bill Of Material(BOM) explosion algolithms which usually takes heavy computation time. In order to improve the effectiveness of the MRP systems, we propose to build a logical BOM database in advance, which reflects the Join operations for the BOM explosion. It reduces the response time for the BOM indented explosion. It reduces the MRP processing time. It also increased main memory utilization.