• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.05a
• /
• pp.877-880
• /
• 2013

정보시스템의 기반이 되는 관계형 데이터베이스에서는 데이터의 양에 따라 성능 차이가 발생한다. 데이터베이스에 관한 여러 가지 기능에 대한 이해가 부족하여 많은 성능 저하 문제를 유발하는데, 그중에 조인 성능문제가 큰 비중을 차지하고 있다. 아주 드문 경우가 아니라면 대부분의 데이터 처리는 하나 이상의 테이블이 필요하기 때문이다. 조인을 정확하게 사용하면 성능 개선에 큰 이점을 가져 올 수 있다. 본 연구는 관계형 데이터베이스 기반의 가장 기본적인 조인방식인 Nested Loops Join 방식을 효율적으로 수행하기 위한 조인순서 선정 및 인덱스 구성에 관한 연구를 하였다. 연구 결과를 평가하기 위해서 SQL Trace을 추출한 후 성능을 비교함으로써 선정된 조인순서가 효율적인 것을 입증하였다. 또한 기존의 응답시간을 기준으로 성능평가방법보다 액세스한 데이터 블록 수를 기준으로 한 성능 평가방법이 더 근본적으로 조인 성능을 개선할 수 있음을 증명하였다. 차후에는 더 복잡한 조인 형태 및 다른 조인방식의 성능개선 방법에 관한 연구를 진행할 것이다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07a
• /
• pp.457-459
• /
• 2005

OBS(Optical Burst Switching) 망에서 우선 순위를 보장해 주기 위해 채널 선정 방식(preemptive channel reservation scheme)이 제안되어 왔다. 채널 선점으로 인해 drop된 버스트의 크기가 OBS 망의 전체 성능에 영향을 끼치는 요소인데도 불구하고 기존의 채널 선점 방식은 선점될 채널 선택 시에 버스트의 채널 사용 시간(elapsed time)만을 고려하고 있다. 따라서 본 논문에서는 각 버스트의 채널 사용 시간과 버스트 길이의 합이 작은 채널을 선택하는 선점 방식으로써 Length-based Smallest Elapsed time Selection(LSE)을 제안한다 시뮬레이션을 통하여 제안하는 LSE 방식이 기존의 SE(Smallest Elapsed time)에 비해서 버스트 손실 성능 면에서 우수함을 알 수 있었다.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.933-940
• /
• 2018

In this paper, we propose a multi-block deep learning structure for improving face recognition rate. The recognition structure of the proposed deep learning consists of three steps: multi-blocking of the input image, multi-block selection by facial feature numerical analysis, and perform deep learning of the selected multi-block. First, the input image is divided into 4 blocks by multi-block. Secondly, in the multi-block selection by feature analysis, the feature values of the quadruple multi-blocks are checked, and only the blocks with many features are selected. The third step is to perform deep learning with the selected multi-block, and the result is obtained as an efficient block with high feature value by performing recognition on the deep learning model in which the selected multi-block part is learned. To evaluate the performance of the proposed deep learning structure, we used CAS-PEAL face database. Experimental results show that the proposed multi-block deep learning structure shows 2.3% higher face recognition rate than the existing deep learning structure.

• Journal of the Society of Disaster Information
• /
• v.9 no.2
• /
• pp.164-177
• /
• 2013

From 2010 years to 2013, the selection method of successful bidder has been changed design suitable lowest method, comprehensive evaluation method(bid price adjustment, design score adjustments, weight standards), fixed amount best design method. Therefore this study was presented by analyzing the impact on the domestic construction industry according the way to determine the successful bidder, the evaluation order method and the operational issues through the analysis of order status, order number, results of acceptance an order, success rate range, results of acceptance an order according to weight success rate, results of acceptance an order by work type and order institution of a turnkey alternative tendering construction performed in domestic civil part.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.8
• /
• pp.1575-1580
• /
• 2017

The electric vehicle business is important because it can reduce 30% of the fine dust generated in the metropolitan area and it can solve the air pollution problem by replacing automobile exhaust gas from an internal combustion engine with eco-friendly electric cars. For the construction of the electric charging station infrastructure, which is the core part of the electric car business, we focus to select the optimal location of the electric car charging station in Seoul. The goal of this paper is to utilize and analyze the traffic statistics of T-Map navigation users data and Seoul Metropolitan Transportation Policy Department to deploy the electric cars charging station with optimal location to increase the efficiency. In this paper, the proposed algorithm is composed of two parts of electric charging station selection. First, we analyze real traffic statistics and area. Second, we utilize T-Map navigation data distribution. To select optimal electric charging station location, we apply these two algorithms.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• 2008.11a
• /
• pp.509-512
• /
• 2008

Clients demands have been changed by construction circumstances such as high-technology, mega-projects, and the complexity of construction. Advanced foreign countries have already introduced and applied new project delivery systems(i.e, CM, DB, etc.) to deal with the change of such that. Recently, Korea also launched two new project delivery systems, "Best Value Contracts to which the State is a Party" to provoke the project delivery system, which helps a public owner select a project delivery system appropriate to the project characteristics. Even though government has made his efforts as stated above, some clients have still used those to public construction projects uniformly. That is because there are not a selection criteria of project delivery system for clients. Therefore, this paper proposes the selection criteria of project delivery system based on project types in the public large construction projects.

• 보일러설비
• /
• s.75
• /
• pp.88-92
• /
• 2000

빙축열시스템의 기존운전방식은 냉동기와 빙축열조의 적절한 조합운전에 의하여 크게 두가지 방식이 이용되고 있다. 이것은 냉동기우선 그리고 빙축열우선 운전방식이며 운전제어의 용이성 그리고 비용측면의 고려 등으로부터 선정되고 있다. 하지만 이러한 운전방식은 운전비용을 절감하기 위한 최선의 기법은 되지 못하고 있으므로, 더욱 운전비를 절감할 수 있는 운전기법을 계획하고 이것으로부터 제어기를 위한 내용에 대하여 설명하고자 한다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.10 no.5
• /
• pp.60-70
• /
• 2011

Vehicle location detection and wireless communication method have be designed along the same lines as GPS, CDMA and WLAN, which is based upon the selecting factors such as state-of-art technology trend, accuracy, stability, and economic feasibility, in order to select the optimum method of information communication networks for integrated "Transportation-Power-Facilities" on the electric bus. In addition, the key features of each alternative for an efficient linkage have been review and the integration methodology for linking among Transportation Charging Center, Transportation(ITS, BIS) Center and smart Grid Center has been drawn up based on a technical level of difficulty of each alternative, political and administrative difficulties, and expense justification.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 2006.07a
• /
• pp.651-651
• /
• 2006

• 한국도로학회지:도로
• /
• v.19 no.1
• /
• pp.48-54
• /
• 2017