Eunmi Choi

Kookmin University, Sŏul, Seoul, South Korea

Are you Eunmi Choi?

Claim your profile

Publications (64)5.53 Total impact

  • Mino Ku, Eunmi Choi, Dugki Min
    [Show abstract] [Hide abstract]
    ABSTRACT: SUMMARY To process big data in a cloud computing environment, a large scale of stream big data is one of the challenging issues. In particular, complex event processing is an emerging technology to handle massive events from multiple sources and process them in a near real time. In this paper, we analyze four performance-influencing factors on a virtualized event processing system: the number of query statements, garbage collection intervals, the quantity of virtual machine resources, and virtual CPU assignment types. In our experiments, we observe the performance effects of these performance parameters, by implementing and running an Esper-based event processing application on top of a Xen-based virtualized system. With experimental results, we analyze the memory consumption problem, and apply periodic garbage collection, to reduce unexpected memory consumption of JVM. Also, we analyze performance effects of the number of cores in a virtual machine (VM) and resource sharing among VMs. Under the virtualized environment in cloud computing infrastructure, one of the critical issues is the management of virtualized computing resources. Accordingly, we present the event processing performance on VMs as a function of virtual CPU assignment types and the number of VMs that share virtual CPUs. Copyright © 2014 John Wiley & Sons, Ltd.
    International Journal of Communication Systems 01/2014; · 1.11 Impact Factor
  • Don S, Duckwon Chung, Eunmi Choi, Dugki Min
    [Show abstract] [Hide abstract]
    ABSTRACT: Real-time remote health monitoring systems are experiencing tremendous advancement resulting from improvements in low power, reliable sensors; yet they are still constrained to low-level interpretation. Automatic data analysis continues to be a tedious task due to a lack of efficient, reliable platforms for data analysis. In this paper, we present a system for monitoring patients remotely by emphasizing the strength of Complex Event Processing (CEP) and Situation Awareness. In this approach, the system makes decisions in a declarative way, which helps medical experts to understand the situation in a more realistic manner. The primary objective of this paper is to explicate the different components inside the system. To verify the technical feasibility of each component, the proposed system is implemented and tested using ECG data.
    Journal of Medical Systems 04/2013; 37(2):9901. · 1.78 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study, we present a three-stage method for detecting abnormalities and classifying electrocardiogram (ECG) beats using a k-nearest neighbor (k-NN) classifier and Gaussian mixture model (GMM). In the first stage, a signal filtering method is used to remove the ECG beat baseline wander. In the second stage, features are extracted based on Higuchi's fractal dimension (HFD) and statistical features. In the third stage, k-NN and GMM are used as classifiers to classify arrhythmia and ischemia. A total of 30,000 ECG segments obtained from the MIT-BIH Arrhythmia and European ST-T Ischemia databases were used to quantify this approach. 60% of the beats were used for training the classifier and the remaining 40%, for validating it. An overall accuracy of 99% and 98.24% was obtained for k-NN and GMM, respectively. This result is significantly better than that of currently used state-of-the-art classification approaches for arrhythmia and ischemia.
    Journal of Mechanics in Medicine and Biology 01/2013; 13(01). · 0.76 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In the past decades there has been substantial research undergone in design of intelligent architecture for Rotorcraft based Unmanned Aerial Vehicle (RUAV). Designing intelligent architecture is a challenging problem since future RUAV’s are utterly autonomous and their performance is comparable to that of manned vehicles. This paper deals with design and development of a layered architectural framework which addresses the issue arising in autonomous intelligent control system. The architecture consists of two layers: the high-level layer occupied by planning routines. In this level the way-points, mission tasks from command center is executed. The low-level layer function is to stabilize the flight and follow the commanded trajectory from upper layer. These layers integrate the following functionalities: (a) way-points navigation and control which includes auto-landing (b) obstacle detection and avoidance, (c) fault detection and identification, and (d) system reconfiguration in two levels (high level and low level controllers). The resulting layered architecture is discussed in detail. More over the novel fault detection and identification method is developed to address multiplicative and additive faults. The testing environment for RUAV is developed to validate our architecture. Complete setup is carried out and run under QNX RTOS, based PC104 embedded board. The algorithms are tested and evaluated using HILS (Hardware-in-the-Loop simulation). Simulation result proves that the proposed architecture demonstrates desired efficiency and reliability.
    Journal of Aerospace Engineering 11/2012; · 0.78 Impact Factor
  • S. Don, Eunmi Choi, Dugki Min
    [Show abstract] [Hide abstract]
    ABSTRACT: With the rapid development in IT enabled health care systems, the monitoring and diagnosis of patient have been replaced by embedded software systems. The limitations of these systems are interoperability and data aggregation across a wide range of devices. To solve these issues, the Medical Cyber Physical Systems (MCPS) has evolved as a new paradigm for health care systems, where the patients will have a flexible interaction with the cyber world in order to achieve an optimal decision result. In this paper we propose an architectural design framework for remote health care monitoring system. This framework is a dynamic event based sensing and monitoring system designed by integrating the full functionality of cyber physical system. The system adapts the concept of local and global awareness for intelligent decision making in physical and cyber world. A test scenario was considered to demonstrate the technical feasibility of the proposed system.
    Awareness Science and Technology (iCAST), 2012 4th International Conference on; 01/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: Cloud computing has emerged as a popular computing milieu that provides a range of delivering solutions for small to large enterprises with a flexible model that allows a computing power and storing space for the large volumetric data within minimum cost. These days, computational paradigm is shifting towards utility‐based pay‐as‐you‐go model and many discussion aside, but there remains no canonical definition of cloud computing yet. In this paper we have proposed a service‐oriented taxonomical spectrum of cloud computing, which is more focused on the service engineering perspective of cloud. Our argument behind cloud engineering is a layered structural approach ‘as a Service’ such as security as a service, fault tolerance as a service, architecture as a service. The main contribution of this paper is to identify a wide spectrum of taxonomy, aiming at a better understanding of functional as well as architectural components that could benefit from cloudification. We describe each sub‐taxonomy (architecture, core services, security, fault tolerance, management services etc.) in details. In addition, we present a comparative study of several cloud systems based on taxonomy. Moreover, it also identifies many challenges and opportunities that exist on the landscape of enterprise cloud. Copyright © 2011 John Wiley & Sons, Ltd.
    International Journal of Communication Systems 01/2012; 25(6). · 1.11 Impact Factor
  • Source
    04/2011; , ISBN: 978-953-307-220-3
  • N.M.K. Varma, Dugki Min, Eunmi Choi
    [Show abstract] [Hide abstract]
    ABSTRACT: In the current trend Virtual Machine Monitors (VMMs) are used as software-based solutions in the enterprise environments for building shared hardware infrastructures via virtualization. Proper utilization of the computer resources is one of the main goals of VMM. Xen is widely used VMM. The main aim of this paper is to diagnose the CPU utilization with different scenarios using the Xen VMM. This paper explains about three kinds of scenarios to diagnose the CPU utilization in Xen virtual machine environment. First scenario is to run the homogeneous virtual machines (VMs) with same memory allocation. In this scenario CPU utilization is diagnosed by adding number of homogeneous VMs to the Xen VMM with equal memory allocation to each VM. Second scenario is to run the homogeneous VMs with different memory allocation. CPU utilization is diagnosed in the second scenario by adding number of homogeneous VMs to the Xen with different sizes of memory allocation to each VM. In the third scenario CPU utilization is diagnosed by running maximum number of homogeneous VMs with same memory allocation in different memory configured systems.
    Computer Sciences and Convergence Information Technology (ICCIT), 2011 6th International Conference on; 01/2011
  • Convergence and Hybrid Information Technology - 5th International Conference, ICHIT 2011, Daejeon, Korea, September 22-24, 2011. Proceedings; 01/2011
  • [Show abstract] [Hide abstract]
    ABSTRACT: We propose a robot teleoperation service architecture that enables us to control robot more reliable. Reliability is one of key issues in the researches of network based robot architectures. Most of researches on robot architecture are to solve network related problems like latency or disconnection by implementing artificial intelligence at robots. In this paper, we explain 3-tier client-server network-based Robot Teleoperation Service Architecture. Users, who control the robot, can use any controllers and robots by this service architecture. Even if users are using two controllers and one robot, Kalman-based Signal Filter is reinforcing and separating these signals constantly. This kind of approach provides control that is more accurate and the result of test systems proves the proposed architecture reliability.
    01/2011;
  • Mino Ku, Dugki Min, Eunmi Choi
    [Show abstract] [Hide abstract]
    ABSTRACT: Virtualization technology is one of the key technology by explosive growth in several computing paradigm including cloud computing. Specially, some virtualization software stacks widely support virtualization mechanism on various system types including a system which doesn't support hardware level virtual- ization. Thus the virtualization technique allows various systems to become a computing platform in virtualized computing envi- ronment. However, some virtualized systems create critical utili- zation by computation power during creation of virtual machine. In our research, we analyze creation characteristics of virtual machine on various dimensions like virtual machine, virtualized system, and storage server. To perform our study, we implement an automated measurement method and test bed which employs Xen as hypervisor on top of the cluster systems with storage serv- er for distributing virtual disk images. Our experiments indicate that computational power of virtualized system affects the boot time of virtual machine in our test bed even the heterogeneity is relatively row. Also, the creation of virtual machine creates sig- nificant overhead (100%) at initial phase on virtualized system in worst case. Additionally, we provide overhead analysis on storage server through generating workloads by scaling virtual machines.
    01/2011;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Efficient traffic control systems are required in towns and cities around the world. As the number of vehicles increase, and new infrastructure is less viable; high levels of mobility become essential. This work proposes a strategy for Intelligent Transportation Systems (ITS), in particular for the task of adaptive signal management. This approach is based on Complex Event Processing (CEP). Event processing provides fast assessment of the situation, due to its inherent capabilities for pattern detection and information composition. This work proposes an architectural model based on the concept of event processing networks. Our model is specifically designed to derive events obtained from loop detectors and compose enhanced signal timings, representing accurately the requirements of the roadway for signal management.
    Networked Computing and Advanced Information Management (NCM), 2010 Sixth International Conference on; 09/2010
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes a classification method based on Complex Event Processing (CEP) for mammogram images. For breast cancer detection and early prevention, several factors must be considered. Mammogram is important and effective in breast cancer diagnosis. The Radiologist must analyze visually, based on his expertise, to determine whether it is cancer or not. We explore the use of CEP for breast cancer image classification and information enrichment. The results reveal that the application of event processing techniques improves the classification of images in the medical domain. The proposed system performs efficiently and timed-lined to help the Radiologist to make a more accurate assessment.
    New Trends in Information Science and Service Science (NISS), 2010 4th International Conference on; 06/2010
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper is to survey video encryption algorithms field which becomes as an important field nowadays. As an increasing rate of applying video is getting high, the security of video data becomes more important. In this paper listed several famous encryption schemes which have their own advantages and cryptographic stability. Three main classes of video encryption techniques are studied. Given encryption which based on entropy coding, reason is entropy coding based encryption solutions are regarded as one of the bests.
    Networked Computing (INC), 2010 6th International Conference on; 06/2010
  • [Show abstract] [Hide abstract]
    ABSTRACT: Secure data transmission and data routing are critical in designing cluster based sensor network. This paper presents a secure energy efficient dynamic routing scheme (SEEDR) for wireless sensor networks. SEEDR uses a symmetric cryptography algorithm to support security. The dynamic key exchange protocol based on DH (Diffie-Hellman) algorithm is proposed, with non blocking OVSF codes. Our analytical model and the security analysis prove that SEEDR increase data communication security and consume less communication energy.
    Innovative Computing and Communication and Asia-Pacific Conference on Information Technology and Ocean Engineering, International Conference on. 01/2010;
  • B.P. Rimal, Eunmi Choi
    [Show abstract] [Hide abstract]
    ABSTRACT: Cloud computing has emerged as a popular computing model to support processing volumetric data using clusters of commodity computers. Nowadays computational world is opting for pay-for-use models and hype and discussion aside, there remains no canonical definition of cloud computing. In this paper we proposed a wide spectrum of taxonomy for cloud computing, aiming at a better understanding of the categories of applications that could benefit from cloudification and that will address the landscape of enterprise IT, management services, and data governance etc.
    Ubiquitous Information Technologies & Applications, 2009. ICUT '09. Proceedings of the 4th International Conference on; 01/2010
  • Source
    B. Upadhyaya, Eunmi Choi
    [Show abstract] [Hide abstract]
    ABSTRACT: Social networks has changed the way we use Web. There is a huge amount of data created and generated by users. There is an obvious question about who owns your data you created. Can you take your data (FOAF information, bookmarks, tags, photos etc) as you leave the service? Browsing, tagging and bookmarking pattern can easily raise the issues of privacy. People are separated over many existing social sites which make it difficult to know the social relationship among the users. Most of the current search engines don't utilize social connection of users to resolve their query. Search results need to be personalized based peoples' relationship and their contribution at the same time we want to ensure the privacy of the users. Much of the information is still restricted to different sites and search is not possible due to the lack of technical and semantic interoperability, locking in knowledge resources in disconnected islands. In this paper we tried to develop an infrastructure that utilizes the relationship between people and their interaction with people to people, people to content. We present an approach of integration of information from different sources bringing together the concept from semantic technology, P2P and social sites. We also propose a novel publish subscribe mechanism for social networks that can benefit from the social relation between peoples to disseminate among the users.
    Ubiquitous Information Technologies & Applications, 2009. ICUT '09. Proceedings of the 4th International Conference on; 01/2010
  • [Show abstract] [Hide abstract]
    ABSTRACT: In Unmanned Aerial Vehicle system, the application software is large and becoming hard to meet the real-time application. UAV system may suffer faults in the controlled plant as well as in the execution platform. The execution platforms support a modern real-time embedded system, but distributed architecture is made of heterogeneous components that may incur transient or permanent faults. In the proposed system, we have developed UAV system based on Hardware in-the Loop Simulation (HILS), which is used for testing the UAV with real time data and real environment. This technique of simulation makes the system to be tested extrinsically with various conditions. The major part of this system deals with fault tolerance system for the UAV with fault injection mechanism and fault detection mechanism using the fault tree analysis. The fault recovery mechanism is also used for making the UAV to land in the safe mode. This paper also deals with the hardware in the simulation setup for the UAV system using the QNX operation system. This reliable architecture can enhance analysis capabilities for critical safety properties and reduce costs for such systems using Hardware in-the Loop Simulation.
    01/2010;
  • [Show abstract] [Hide abstract]
    ABSTRACT: User Created Robots (UCRs) are personal robots that consist of toy blocks. Usually user controls UCR via short-range wireless communications such as IR, ZigBee, Bluetooth and so on. However, this current technology has some spatial limitations. UCR remote control system gives extension of UCR related services and activities. Also this system allows in conjunction with e-education services and other external services such as web services and virtual worlds. In this paper, we handle two key issues for expansion of UCR applications. One is how to control UCR in a place far away, and the other is how to support interoperability between UCR and its controller. We solve these issues by network protocol transformation and logical key mapping structure. We suggest system architecture of UCR remote control system that supports interoperability and our implementation and evaluation results show efficiency of proposed system.
    01/2010;
  • Mino Ku, Dugki Min, Eunmi Choi
    [Show abstract] [Hide abstract]
    ABSTRACT: Cluster computing has been widely used in various fields from academics to industry because of its technical advantages. Recently, cloud computing become more attractive, as cluster computing is one of the core basements of cloud computing. However, the development of cluster-based system is hard without relative knowledge such as networking, workload distribution, fault tolerant, etc. Thus, various implementations for cluster computing have been introduced to help the development of cluster-based system. However, these implementations also have the limitation for developing cluster-based system because of high performance in specified application fields. In this paper, we present the SCAREX framework which supports scalability, reliability, extendibility, and easy for implementing cluster-based system. This framework employs a fault-tolerant mechanism for the failure of workload processing, and facilitates dynamic module distribution and loading for the work module which works on the worker node. The SCAREX also provides a simple programming interface for easy development. Since the SCAREX has flexible architecture, the developer can implement various cluster architecture and extend cluster architecture by combining the SCAREX-based cluster with other packages of the SCAREX or other frameworks.
    01/2010;

Publication Stats

275 Citations
5.53 Total Impact Points

Institutions

  • 2004–2014
    • Kookmin University
      • School of Management Information Systems
      Sŏul, Seoul, South Korea
    • Handong Global University
      Geijitsu, North Gyeongsang, South Korea
  • 2008–2013
    • Konkuk University
      • Division of Computer Science and Engineering
      Seoul, Seoul, South Korea
  • 1996
    • Michigan State University
      • Department of Computer Science and Engineering
      East Lansing, MI, United States