• KIISE Transactions on Computing Practices
• /
• v.23 no.4
• /
• pp.199-209
• /
• 2017

Recently, there has been growing interest in workload consolidation via virtualization for real-time systems. Meanwhile, due to virtualization, additional overhead from intervention of hypervisor increases the execution time of applications running on virtual machine. The increase in execution time decreases the performance of workloads on virtual machines, thus satisfying real-time requirements are not easy. To resolve the problem, we designed and implemented a hypervisor (the K-Hypervisor) that allows programs on virtual machines to access the physical resources, without intervention of the hypervisor. Experimental results show that programs running on K-Hypervisor increase their execution time by about 3% on an average, compared to the native environment. Moreover, it is suitable for real-time workload execution because of its uniform performance degradation, regardless of the resources accessed from tasks, and the frequency it is accessed.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.116-120
• /
• 2011

In this paper, we have designed and implemented the integrated protocol gateway for the integration of CAN and Profibus networks. To do that, we used the Switched Ethernet as Backbone network, and each fieldbus network is connected by CAN/Ethernet or Profibus/Ethernet gateway, and each gateway perform the translation between fieldbus and Ethernet frames. Futhermore, we realized the real-time features in the environment of the Switched Ethernet by applying the distributed hard real-time scheduling algorithm among each gateways. To implement tht CAN/Ethernet and Profibus/Ethernet gateways, we used the Linux of kernel 2.6.31.12 real-time patched version(PREEMTED_RT), and we could verify successful message translation and real-time features through real implementation.

• Journal of KIISE:Computing Practices and Letters
• /
• v.13 no.5
• /
• pp.271-281
• /
• 2007

Real-time support of embedded OS is not optional, but essential in contemporary embedded systems. In order to achieve these system#s real-time property, it is crucial that schedulability analysis for tasks having its property have been accomplished before system execution. Acquiring Worst-Case Execution Time(WCET) of task is a core part of schedulability analysis. Because traditional WCET tools analyze only its estimation of application task(i.e. program), it is not considered that application tasks are affected by scheduling primitives(e.g. scheduler, interrupt service routine, etc.) of OS when it schedules them. In this paper, we design and implement WCET analysis tool which deliberates on scheduling primitives of system using embedded Linux widely used in embedded OSes. This tool can estimate either WCET of normal application programs or corresponding primitives which have an influence on schduling property in embedded Linux kernel. Therefore, precision of estimation about schedulability analysis is improved. We develop this tool as Eclipse#s plug-in to work properly in any platform and support convenient interface or functionality for user.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.591-596
• /
• 2019

In this paper, we implemented a deep learning operation structure with less influence of local PC performance. In general, the deep learning model has a large amount of computation and is heavily influenced by the performance of the processing PC. In this paper, we implemented deep learning operation using AWS and streaming server to reduce this limitation. First, deep learning operations were performed on AWS so that deep learning operation would work even if the performance of the local PC decreased. However, with AWS, the output is less real-time relative to the input when computed. Second, we use streaming server to increase the real-time of deep learning model. If the streaming server is not used, the real-time performance is poor because the images must be processed one by one or by stacking the images. We used the YOLO v3 model as a deep learning model for performance comparison experiments, and compared the performance of local PCs with instances of AWS and GTX1080, a high-performance GPU. The simulation results show that the test time per image is 0.023444 seconds when using the p3 instance of AWS, which is similar to the test time per image of 0.027099 seconds on a local PC with the high-performance GPU GTX1080.

• Journal of Digital Convergence
• /
• v.17 no.8
• /
• pp.283-291
• /
• 2019

The aim of investigated the role of FABP5 in the hepatic lipogenesis and lipid metabolisms. Mice were overexpressed and silenced liver FABP5 using virus particles. Mice were fed a Western-type diet or regular chow for 1week and then sacrificed mouse after 24hr fasted. Liver homogenates were used for protein analysis by Western blot and mRNA levels by RT-PCR. Hepatic and serum lipids were analysed by thin-layer chromatography. Mice fed a Western-type or high saturated fat diet revealed large increases in FABP5 expression. However, FABP5 mRNA levels were drastically reduced under fasted. Hepatic TG was significantly increased FABP5-OEAV mice, but a significantly decreased hepatic free cholesterol under fed. The discovered a substantial decrease in hepatic TG mass with FABP5 silencing. In these data, presented evidence for an important role of FABP5 in hepatic lipogenesis and hepatic TG storage. FABP5 may also be a potential target in the treatment of NAFLD, metabolic syndrome, and obesity. Furthermore, studies to which transcription factors are involved in FABP5 expression and regulation.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.54-57
• /
• 2019

모바일 디바이스와 스마트 글래스의 확산/보급으로 다양한 AR/VR/MR 어플리케이션이 출시 되었으나, 사용자들이 일반적으로 가정에서 많이 사용하는 대형 Display인 TV에서 이용할 수 있는 AR/VR/MR 어플리케이션은 거의 없는 실정이다. TV 디바이스에서 이러한 영상처리 기법을 이용하려면 별도의 카메라나 CPU가 필요한데, 이를 이용하기 위해 사용자들이 별도의 투자를 하기 어렵기 때문이다. 본 논문에서는 사용자에게 가장 친숙한 모바일 디바이스와, 디지털 TV신호를 수신하기 위한 STB를 연동하여 다양한 AR/VR/MR 서비스를 이용할 수 있는 방법을 제안하고 실제 시스템 구현과 실험을 통해 제안된 방법의 품질 및 실시간성 향상을 검증한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10e
• /
• pp.406-408
• /
• 2002

TFRC 혼잡제어는 멀티미디어, 인터넷 전화등 실시간성이 강조된 서비스에 적합한 프로토콜이다. 그러나 이러한 TCP-friendly 혼잡제어 알고리즘은 네트워크 혼잡이 심해져 손실률이 커짐에 따라 TCP와의 공정성(fairness)이 떨어지는 문제점을 가지고 있다. 이러한 문제를 해결하기 위해서는 전송률을 결정하는 제어식을 높은 손실률일 때 보정을 해주어야 한다. 본 논문에서는 TFRC 혼잡제어 알고리즘의 전송률 결정에 있어서 중요한 변수인 재전송 타임아웃 값(RTO)의 재 정의를 통하여 전송률을 보정함으로서 공정성 하락을 방지함을 보이고, 시뮬레이션을 통하여 이를 입증한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.134-135
• /
• 2016

채널이 혼잡한 네트워크에서 응용이 요구하는 QoS를 보장하려면 무선 자원의 효율적인 분배가 필요하다. 그러나 기존 802.11의 MAC 프로토콜로는 응용의 다양한 요구사항을 보장하기가 어렵다. 특히 혼잡한 네트워크 환경에서는 전송 지연이 증가하고 데이터 손실이 생긴다. 따라서 본 논문에서는 응용이 요구하는 전송 지연과 실시간성을 충족시킬 수 있도록 경쟁 윈도우 크기를 조절하는 매체 접근 기법을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.11a
• /
• pp.727-730
• /
• 2005

최근 로봇에 대한 연구가 활발히 진행되면서 로봇이 움직이면서 필요한 거리정보를 얻기 위한 여러 가지 3D Sensing 알고리즘이 제시되었다. 그 중 하나인 HOC(hierarchical orthogonal code) 알고리즘 이용하면서 실시간성 및 모듈화를 위해 CIS(CMOS Image Sensor), FPGA 를 이용하여 구현 하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1995.06a
• /
• pp.35-39
• /
• 1995

멀티미디어 데이터는 크기가 크고 실시간성의 특징을 가지고 있다. 특히 HDTV 서버는 많은 수의 비디오를 저장하기 위해서 대용량의 저장 능력이 필요하며 동시에 많은 수의 사용자를 지워하기 위해서 높은 전송율을 보장하여야 한다. 따라서 HDTV 서버는 확장성 있는 하드웨어 구조를 가져야 하고 데이터를 효과적으로 저장 관리할 수 있는 소프트웨어가 지원되어야 한다. 본 논문에서는 HDTV의 비디오 데이터를 저장 관리하기 위한 확장성 있는 HDTV 저장 서버 구조, 고성능 입출력시스템에서 기본적으로 사용하는 데이터 분할 방법, 데이터 배치 방법, 캐싱 정책, admission control 방법과 성능 분석에 대한 내용을 기술한다.