• Journal of KIISE:Databases
• /
• v.37 no.3
• /
• pp.146-164
• /
• 2010

Due to an increasing volume of XML data, it is considered prudent to store XML data on an industry-strength database system instead of relying on a domain specific application or a file system. For shredded XML data stored in relational tables, however, it may not be straightforward to apply existing algorithms for twig query processing, since most of the algorithms require XML data to be accessed in a form of streams of elements grouped by their tags and sorted in a particular order. In order to support XML query processing within the common framework of relational database systems, we first propose several bitmap indexes and their strategies for supporting holistic twig joining on XML data stored in relational tables. Since bitmap indexes are well supported in most of the commercial and open-source database systems, the proposed bitmapped indexes and twig query processing strategies can be incorporated into relational query processing framework with more ease. The proposed query processing strategies are efficient in terms of both time and space, because the compressed bitmap indexes stay compressed during data access. In addition, we propose a hybrid index which computes twig query solutions with only bit-vectors, without accessing labeled XML elements stored in the relational tables.

• KIPS Transactions on Computer and Communication Systems
• /
• v.5 no.12
• /
• pp.455-462
• /
• 2016

Wireless sensor networks have been researched to gather data about events on sensor fields from sources at sinks. Multipath routing is one of attractive approaches to reliably send data against the problem of frequent breakages on paths from sources to sinks due to node and link failures. As mobile events such as humans, animals, and vehicles are considered, sources may be continuously generated according to the movement of the mobile event. Thus, mobile events provide new challenging issue in multipath routing. However, the research on multipath routing mainly focus on both efficient multipath construction from sources to static sinks and fast multipath reconstruction against path breakages. Accordingly, the previous multipath routing protocols request each source continuously generated by a mobile event to construct individual multipath from the source to sinks. This induces the increase of multipath construction cost in the previous protocols in proportion to the number of source. Therefore, we propose efficient multipath routing protocol for supporting continuous sources generated by mobile events. In the proposed protocol, new source efficiently reconstructs its multipath by exploiting the existing multipath of previous sources. To do this, the proposed protocol selects one among three reconstruction methods: a local reconstruction, a global partial one, and a global full one. For a selection decision, we provide an analytical energy consumption cost model that calculates the summation of both the multipath reconstruction cost and the data forwarding cost. Simulation results show that the proposed protocol has better performance than the previous protocol to provide multipath routing for mobile events.

• The KIPS Transactions:PartB
• /
• v.9B no.5
• /
• pp.625-634
• /
• 2002

In this paper, we conducted various experiments with Bayesian networks in order to analyze clinical data of infertility patients. With these experiments, we tried to find out inter-dependencies among important factors playing the key role in clinical pregnancy, and to compare 3 different kinds of Bayesian network classifiers (including NBN, BAN, GBN) in terms of classification performance. As a result of experiments, we found the fact that the most important features playing the key role in clinical pregnancy (Clin) are indication (IND), stimulation, age of female partner (FA), number of ova (ICT), and use of Wallace (ETM), and then discovered inter-dependencies among these features. And we made sure that BAN and GBN, which are more general Bayesian network classifiers permitting inter-dependencies among features, show higher performance than NBN. By comparing Bayesian classifiers based on probabilistic representation and reasoning with other classifiers such as decision trees and k-nearest neighbor methods, we found that the former show higher performance than the latter due to inherent characteristics of clinical domain. finally, we suggested a feature reduction method in which all features except only some ones within Markov blanket of the class node are removed, and investigated by experiments whether such feature reduction can increase the performance of Bayesian classifiers.

• Journal of Korea Spatial Information System Society
• /
• v.10 no.3
• /
• pp.79-91
• /
• 2008

Recently, with highly Interest In Location-Based Service(LBS) utilizing location data of moving objects, the GALIS(Gracefully Aging Location Information System) which is a cluster-based distributed computing architecture was proposed as a more efficient location management system of moving objects. In the SLDS(Short-term location Data Subsystem) which Is a subsystem of the GALIS, since the SDP(Short-term Data Processor) Master transmits current location data and queries to every SDP Worker, the SDP Master reassembles and sends query results produced by SDP Workers to the client. However, the services are suspended during the SDP Master under failure and the response time to the client is increased if the load is concentrated on the SDP Master. Therefore, in this paper, the extended SLDS was designed and implemented to solve these problems. Though one SDP Master is under failure, the other can provide the services continually, and so the extended SLDS can guarantee the high reliability of the SLDS. The extended SLDS also can reduce the response time to the client by enabling two SDP Masters to perform the distributed query processing. Finally, we proved high reliability and high availability of the extended SLDS by implementing the current location data storage, query processing, and failure takeover scenarios. We also verified that the extended SLDS is more efficient than the original SLDS through the query processing performance evaluation.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10a
• /
• pp.414-418
• /
• 2006

유비쿼터스 컴퓨팅의 인프라가 되는 센서 네트워크는 매우 작은 하드웨어로 이루어지는 많은 수의 센서 노드들로 구성된다. 이 네트워크의 토폴로지와 라우팅 방식은 그 목적에 따라 결정되어야 하며, 하드웨어 및 소프트웨어도 필요한 경우에는 변경되어야 한다. 따라서 그러한 네트워크를 최적으로 설계하기 위해서는 시스템 동작을 확인하고 성능을 예측할 수 있는 센서 네트워크 시뮬레이터가 필요하다. 현존하는 몇몇 센서 네트워크 시뮬레이터들은 특정 하드웨어나 운영체제에 맞추어 개발되었기 때문에, 그러한 특정 시스템들을 위해서만 사용될 수 있다. 그리고 시스템 설계 상의 주요 이슈가 되는 전력 소모량 및 프로그램 실행 시간을 추정하기 위한 어떤 수단도 지원하지 못하고 있다. 이 연구에서는 기존의 센서 네트워크 시뮬레이터들이 갖고 있는 문제점을 해결한 시뮬레이터를 개발하고, 센서 네트워크의 계층적 라우팅 프로토콜인 LEACH, TEEN, APTEEN의 전력소모량을 시뮬레이션을 이용하여 분석하였다. 시뮬레이션의 작업부하인 명령어 트레이스로는 ATmega128L 마이크로컨트롤러용 크로스컴파일러에 의해 생성된 실행 이미지를 사용하였다. 따라서 각각의 라우팅 프로토콜을 실제 센서 보드에서 동작하는 응용 프로그램으로 구현하고, 컴파일된 실행 이미지를 작업부하로 사용하여 시뮬레이션 하였다. 라우팅 프로그램들은 ETRI의 센서 네트워크 운영체제인 Nano-Q+ 1.6.1을 기반으로 구현되었으며, 하드웨어 플랫폼은 옥타컴의 센서 보드인 Nano-24이다. 시뮬레이션 결과에 따르면, 센서 네트워크는 그 사용 목적에 따라 라우팅 프로토콜을 적절히 선택해야 한다는 것을 알 수 있다. 즉, LEACH는 주기적으로 네트워크의 상황을 체크해야 하는 경우에 적합하고, TEEN은 환경의 변화를 수시로 감지해야 하는 경우에 적합하다. 그리고 APTEEN은 전력소모량과 기능 측면을 모두 고려할 때 가장 효과적인 라우팅 프로토콜이라고 할 수 있다.iRNA 상의 의존관계를 분석할 수 있었다.수안보 등 지역에서 나타난다 이러한 이상대 주변에는 대개 온천이 발달되어 있었거나 새로 개발되어 있는 곳이다. 온천에 이용하고 있는 시추공의 자료는 배제하였으나 온천이응으로 직접적으로 영향을 받지 않은 시추공의 자료는 사용하였다 이러한 온천 주변 지역이라 하더라도 실제는 온천의 pumping 으로 인한 대류현상으로 주변 일대의 온도를 올려놓았기 때문에 비교적 높은 지열류량 값을 보인다. 한편 한반도 남동부 일대는 이번 추가된 자료에 의해 새로운 지열류량 분포 변화가 나타났다 강원 북부 오색온천지역 부근에서 높은 지열류량 분포를 보이며 또한 우리나라 대단층 중의 하나인 양산단층과 같은 방향으로 발달한 밀양단층, 모량단층, 동래단층 등 주변부로 NNE-SSW 방향의 지열류량 이상대가 발달한다. 이것으로 볼 때 지열류량은 지질구조와 무관하지 않음을 파악할 수 있다. 특히 이러한 단층대 주변은 지열수의 순환이 깊은 심도까지 가능하므로 이러한 대류현상으로 지표부근까지 높은 지온 전달이 되어 나타나는 것으로 판단된다.의 안정된 방사성표지효율을 보였다. $^{99m}Tc$-transferrin을 이용한 감염영상을 성공적으로 얻을 수 있었으며, $^{67}Ga$-citrate 영상과 비교하여 더 빠른 시간 안에 우수한 영상을 얻을 수 있었다. 그러므로 $^{99m}Tc$-transierrin이 감염 병소의 영상진단에 사용될 수 있을 것으로 기대된다.리를 정량화 하였다. 특히 선조체에서의 도파민 유리에 의한 수용체 결합능의 감소는 흡연에 의한 혈중 니코틴의 축적 농도와 양의 상관관계를 보였다(rho=0.9, p=0.04). 결론: $[^{11}C]raclopride$ PET을 이용하여 비흡연 정상인에서 흡연에 의한 도파민 유리를 영상화 및 정량화 하였고, 흡연에 의한 선조체내 도파민 유리는 흡연시 흡수된

• Journal of KIISE:Databases
• /
• v.36 no.6
• /
• pp.476-485
• /
• 2009

In sensor networks, many methods have been proposed to process in-network aggregation effectively. Unlike general aggregation queries, skyline query processing compares multi-dimensional data for the result. Therefore, it is very difficult to process the skyline queries in sensor networks. It is important to filter unnecessary data for energy-efficient skyline query processing. Existing approach like MFTAC restricts unnecessary data transitions by deploying filters to whole sensors. However, network lifetime is reduced by energy consumption for many false positive data and filters transmission. In this paper, we propose a bottom up filtering-based skyline query processing algorithm of in-network for reducing energy consumption by filters transmission and a PBFiltering technique for improving performance of filtering. The proposed algorithm creates the skyline filter table (SFT) in the data gathering process which sends from sensor nodes to the base station and filters out unnecessary transmissions using it. The experimental results show that our algorithm reduces false positives and improves the network lifetime over the existing method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.12
• /
• pp.2366-2373
• /
• 2007

In this paper, a low-power and small-area asynchronous 1 kilobit EEPROM for passive UHF RFID tag chips is designed with $0.18{\mu}m$ EEPROM cells. As small area solutions, command and address buffers are removed since we design asynchronous I/O interface and data output buffer is also removed by using separate I/O. To supply stably high voltages VPP and VPPL used in the cell array from low voltage VDD, Dickson charge pump is designed with schottky diodes instead of a PN junction diodes. On that account, we can decrease the number of stages of the charge pump, which can decrease layout area of charge pump. As a low-power solution, we can reduce write current by using the proposed VPPL power switching circuit which selects each needed voltage at either program or write mode. A test chip of asynchronous 1 kilobit EEPROM is fabricated, and its layout area is $554.8{\times}306.9{\mu}m2$., 11% smaller than its synchronous counterpart.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.581-584
• /
• 2009

Recently, the cameras which used for observation are installed in children protection area and local crime prevention area in order to protect life and property and by its work being recognized and are installed more. Normal cameras have cost problem to observe multiple area and detail, because they can observe only one place. PTZ camera can observe multiple area by moving focus by schedule or remote control, but it can't automatically move the focus of it to the place where event occurred, because it can't recognize the place. In this study, we can monitor multiple area effectively, by installing a wireless sensor node equipped with temperature, lighting, gas and human detection sensor to each area, to monitor many place low-price and actively and to move the focus of PTZ camera to preset position, and send recorded video to the user, when the various sensor data received from wireless sensors in observation area are to be determined abnormal by analyzing. In addition, at night we can record a scene using infrared, but to reduce power consumption of lighting system which are installed to improve resolution, it supplies power to the lighting system when event occurred. So we were able to implement low power green monitoring system.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.7
• /
• pp.191-198
• /
• 2021

Since the size of training dataset become large and the model is getting deeper to achieve high accuracy in deep learning, the deep neural network training requires a lot of computation and it takes too much time with a single node. Therefore, distributed deep learning is proposed to reduce the training time by distributing computation across multiple nodes. In this study, we propose hybrid allreduce strategy that considers the characteristics of each layer and communication and computational overlapping technique for synchronization of distributed deep learning. Since the convolution layer has fewer parameters than the fully-connected layer as well as it is located at the upper, only short overlapping time is allowed. Thus, butterfly allreduce is used to synchronize the convolution layer. On the other hand, fully-connecter layer is synchronized using ring all-reduce. The empirical experiment results on PyTorch with our proposed scheme shows that the proposed method reduced the training time by up to 33% compared to the baseline PyTorch.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.5
• /
• pp.100-111
• /
• 2019

The existing selected station analysis (SSA) method analyzes the link transfer mode data between origin and destination of individuals passing through stations from a microscopic standpoint. As such, existing SSA is insufficient as it uses integrated analysis using macroscopic data such as subway lines. This research builds a line chain OD based on path search of individual passenger's movement through the subway, and explores means to utilize the findings. First, a method is proposed that searches the traversed subway path from the linked passage modes that the passenger uses and applies the results to SSA line analysis. Compared to the existing SSA, this method provides for analysis of commonly conflicting features such as the line on which the station is passed, and the stations included on the line thanks to the presence of complete information of the individual passenger's traversed path. It also allows for integrated observation of the line chain OD that approaches a certain station. For enhanced understanding, Seoul Metro Line 9 is used as a case study to demonstrate the integrated formulation concept of line chain OD centered around a certain station as well as the macroscopic features of the traversed path that approaches stations included on the line.