• Journal of the Korean Data and Information Science Society
• /
• 제18권4호
• /
• pp.915-925
• /
• 2007

Multivariate GARCH has been useful to model dynamic relationships between volatilities arising from each component series of multivariate time series. Methodologies including EWMA(Exponentially weighted moving-average model), DVEC(Diagonal VEC model), BEKK and CCC(Constant conditional correlation model) models are comparatively reviewed for bivariate time series. In addition, these models are applied to evaluate VaR(Value at Risk) and to construct joint prediction region. To illustrate, bivariate stock prices data consisting of Samsung Electronics and LG Electronics are analysed.

• 전자공학회논문지A
• /
• 제28A권1호
• /
• pp.52-59
• /
• 1991

An effective and accurate delay time model is the key problem in the simulation and timing verification of CMOS logic circuits. We propose a semi-analytic CMOW delay time model taking into account the configuration ratio, the input waveform slope and the load capacitance. This model is based on the Schichman Hodges's DC equations and derived on the optimally weighted switching peak current. The parameters necessary for the model calculation are automatically determined from the program. The proposed model is computationally effective and the error is typically within 10% of the SPICEA results. Compared to the table RC model, the accuracy is inproved over two times in average.

• ETRI Journal
• /
• 제29권3호
• /
• pp.270-280
• /
• 2007

In Linux, real-time tasks are supported by separating real-time task priorities from non-real-time task priorities. However, this separation of priority ranges may not be effective when real-time tasks make the system calls that are taken care of by the kernel threads. Thus, Linux is considered a soft real-time system. Moreover, kernel threads are configured to have static priorities for throughputs. The static assignment of priorities to kernel threads causes trouble for real-time tasks when real-time tasks require kernel threads to be invoked to handle the system calls because kernel threads do not discriminate between real-time and non-real-time tasks. We present a dynamic kernel thread scheduling mechanism with weighted average priority inheritance protocol (PIP), a variation of the PIP. The scheduling algorithm assigns proper priorities to kernel threads at runtime by monitoring the activities of user-level real-time tasks. Experimental results show that the algorithms can greatly improve the unexpected execution latency of real-time tasks.

• 융합보안논문지
• /
• 제11권3호
• /
• pp.19-24
• /
• 2011

소프트웨어 고장 시간은 테스팅 시간과 관계없이 일정하거나, 단조증가 혹은 단조 감소 추세를 가지고 있다. 이러한 소프트웨어 신뢰모형들을 분석하기 위한 자료척도로 자료에 대한 추세 검정이 개발되어 있다. 추세 분석에는 산술평균 검정과 라플라스 추세 검정 등이 있다. 추세분석들은 전체적인 자료의 개요의 정보만 제공한다. 본 논문에서는 고장시간을 측정하다가 시간 절단이 될 경우에 미래의 고장 시간 예측에 관하여 연구 하였다. 시계열 분석에 이용되는 단순이동 평균법과 가중이동평균법, 지수평활법을 이용하여 미래고장 시간을 예측하여 비교하고자 한다. 실증분석에서는 고장간격 자료를 이용하여 모형들에 대한 예측값을 평균자승오차를 이용하여 비교하고 효율적 모형을 선택 하였다.

• Industrial Engineering and Management Systems
• /
• 제4권1호
• /
• pp.23-35
• /
• 2005

This paper aims to develop a practical finite capacity MRP (FCMRP) system based on the needs of an automotive parts manufacturing company in Thailand. The approach includes a linear goal programming model to determine the optimal start time of each operation to minimize the sum of penalty points incurred by exceeding the goals of total earliness, total tardiness, and average flow-time considering the finite capacity of all work centers and precedence of operations. Important factors of the proposed FCMRP system are penalty weights and dispatching rules. Effects of these factors on the performance measures are statistically analyzed based on a real situation of an auto-part factory. Statistical results show that the dispatching rules and penalty weights have significant effects on the performance measures. The proposed FCMRP system offers a good tradeoff between conflicting performance measures and results in the best weighted average performance measures when compared to conventional forward and forward-backward finite capacity scheduling systems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권3호
• /
• pp.894-907
• /
• 2022

The features extracted by convolutional neural networks are more descriptive of images than traditional features, and their convolutional layers are more suitable for retrieving images than are fully connected layers. The convolutional layer features will consume considerable time and memory if used directly to match an image. Therefore, this paper proposes a feature weighting and region integration method for convolutional layer features to form global feature vectors and subsequently use them for image matching. First, the 3D feature of the last convolutional layer is extracted, and the convolutional feature is subsequently weighted again to highlight the edge information and position information of the image. Next, we integrate several regional eigenvectors that are processed by sliding windows into a global eigenvector. Finally, the initial ranking of the retrieval is obtained by measuring the similarity of the query image and the test image using the cosine distance, and the final mean Average Precision (mAP) is obtained by using the extended query method for rearrangement. We conduct experiments using the Oxford5k and Paris6k datasets and their extended datasets, Paris106k and Oxford105k. These experimental results indicate that the global feature extracted by the new method can better describe an image.

• International Journal of Computer Science & Network Security
• /
• 제22권1호
• /
• pp.139-148
• /
• 2022

Research into wireless sensor networks (WSNs) is a trendy issue with a wide range of applications. With hundreds to thousands of nodes, most wireless sensor networks interact with each other through radio waves. Limited computational power, storage, battery, and transmission bandwidth are some of the obstacles in designing WSNs. Clustering and routing procedures have been proposed to address these concerns. The wireless sensor network's most complex and vital duty is routing. With the Greedy Perimeter Stateless Routing method (GPSR), an efficient and responsive routing protocol is built. In packet forwarding, the nodes' locations are taken into account while making choices. In order to send a message, the GPSR always takes the shortest route between the source and destination nodes. Weighted directed graphs may be constructed utilising four distinct distance metrics, such as Euclidean, city block, cosine, and correlation distances, in this study. NS-2 has been used for a thorough simulation. Additionally, the GPSR's performance with various distance metrics is evaluated and verified. When compared to alternative distance measures, the proposed GPSR with correlation distance performs better in terms of packet delivery ratio, throughput, routing overhead and average stability time of the cluster head.

• 한국컴퓨터정보학회논문지
• /
• 제14권2호
• /
• pp.191-199
• /
• 2009

본 논문에서는 핸드폰, 스마트폰, PDA폰 등의 모바일폰에서 동작하는 강건하고 경량화된 한국어 형태소 분석기를 제안한다. 이들 모바일 장치들은 낮은 CPU 성능과 메모리 사용에서의 제약으로 인해 자연언어 인터페이스를 적용하기 곤란했다. 본 논문에서는 1) 키 이벤트 핸들러 (Key Event Handler)를 이용한 온라인 형태소 분석과 2) 붙여 쓴 오류 문장에 대해 띄어쓰기 오류 교정의 전처리 과정 없이 강건한 형태소 분석 방법을 제안한다. 본 논문에서 제안된 경량화된 한국어 형태소 분석기는 모바일 한영 번역기 시제품에 적용하여 메모리 사용량은 5.8% 줄이고 평균 반환 시간은 19.0% 개선하였다.

• 물과 미래
• /
• 제28권1호
• /
• pp.133-144
• /
• 1995

본 연구의 목적은 실시간 자료수집 시스템의 일환으로서 어떤 유역에 존재하는 기존의 강우관측지점으로로부터 유역평균강우량을 산정할 때나 이를 강우-유출 모형의 입력치로 사용할 때 필요한 강우관측지점의 수와 위치를 선정하는 방법을 제시하였다. 유역평균강우량 산정방법은 미국 국립기상청이 제안한 가중치방법을 사용하였다. 강우관측지점 수에 따른 국부적인 최적위치선정은 서로 다른 두가지 방법으로 추적되었다. 유량예측을 위한 강우관측지점의 최적위치와 수를 결정하기 위하여 실제 실무에서 적용하고 있는 강우-유출 모형을 사용하였다.

• Communications for Statistical Applications and Methods
• /
• 제14권1호
• /
• pp.155-167
• /
• 2007

Knowing the time of the process change could lead to quicker identification of the responsible special cause and less process down time, and it could help to reduce the probability of incorrectly identifying the special cause. In this paper, we propose the maximum likelihood estimator (MLE) for the process change point when a control chart is used in monitoring the mean of a process in which the observations can be modeled as an AR(1) process plus an additional random error. The performance of the proposed MLE is compared to the performance of the built-in estimator when they are used in EWMA charts based on the residuals. The results show that the proposed MLE provides good performance in terms of both accuracy and precision of the estimator.