• Journal of Broadcast Engineering
• /
• v.19 no.1
• /
• pp.14-23
• /
• 2014

This paper proposed controlling methods for SDN(Software Defined Network) based multiple radio access technology as the solutions of following two issues which were mainly occurred by explosive increasing of video traffic. The first one is a requirement for traffic off-loading caused by 3rd-party video service providers from the mobile network operator's viewpoint. The other one is a provision of high-speed video contents transmission services with low price. Furthermore, the performance evaluation was also conducted on the real test-bed which is composed of OpenStack cloud and SDN technology such as OpenFlow and Open vSwitch. A virtual machine running on the OpenStack provide a video service and the terminal which is able to use multiple radio access technology supports two 2.4GHz WLANs(Wireless Local Area Network) and three 5GHz WLANs, concurrently. Finally, we can get 820Mbps of the maximum transmission speed by using that five WLAN links for the single service at the same time.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.237-240
• /
• 2003

Efficient memory management is one of the most crucial requirement of lava virtual machine. In Java, memory is allocated everytime when a new instance of class is created or when a method is called. The allocated memory is freed when the instance is no longer used, or when the called method is returned. In this paper we have examined the memory management scheme applied to the simpleRTJ lava virtual machine. The simpleRTJ has such a distinguished characteristic in its memory management scheme that the size of all instances are forced to be the same and the size of stack frames of all methods be the same, respectively. We present in this paper the scheme thoroughly and analyze its anticipated performance qualitatively.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.5
• /
• pp.1058-1067
• /
• 2004

Java virtual machine provides code compactness, simple execution engines, and platform-independence which are important features for small devices such as mobile or embedded device, but it has a big problem, such as low throughput due to stack-oriented operation. In this paper hardware lava accelerator targeted for mobile or embedded application is designed to eliminate the slow speed problem of lava virtual machine. The designed lava accelerator can execute 81 instructions of Java virtual machine(JVM)'s opcodes and be used as Java coprocessor of conventional 32-bit RISC processor with efficient coprocessor interface and instruction buffer. It consists of about 14,300 gates and its maximum operating frequency is about 50 Mhz under 0.35um CMOS technology.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.157-159
• /
• 1998

쓰레드는 프로그램의 동시성을 표현하는데 적합하고, 프로세서 모델에 비하여 동기화나 문맥교환의 비용을 줄일 수 있어 기존의 멀티 프로세스 프로그래밍을 대체하고 있다. 운영체계에서 멀티쓰레팅 환경의 제공은 이제 필수적이며, 좋은 성능을 위해서는 운영체계의 지원이 필요하다. ARX 실시간 운영체계는 유저 레벨 멀티쓰레팅을 지원하고 있으며 쓰레드의 성능을 높이고 유저 레벨에서 실시간 스케쥴링이 가능하도록 하기 위하여 동적 가상 쓰레드 바인딩(dynamic virtual stack binding)과 스케쥴링 이벤트 업콜(scheduling event upcall)등의 기법을 지원한다. 본 논문에서는 자바 가상 머신을 통하여 ARX 운영체계의 쓰레드 라이브러리의 성능을 측정하고 다른 운영체계의 멀티쓰레드 라이브러리와 성능 비교를 하였다. 실험결과 ARX 가 제공하는 쓰레드 라이브러리가 다른 운영체계에 비해 우수한 성능을 보여줌을 확인하였다.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.13 no.5
• /
• pp.75-80
• /
• 2004

Feasibility of the small capacity fuel cell powered vehicle is carried out for system design with loading characteristics. The major design concepts which include battery, driving motor, and fuel cell module are analyzed and discussed for the future development. A load characteristics program is developed in order to calculate the traction power of fuel cell vehicle according to the driving courses specified. Further, the small capacity fuel cell vehicle is analyzed to determine the capacity of stack as a function of the velocity for an appropriate power required.

• 제어로봇시스템학회:학술대회논문집
• /
• 1999.10a
• /
• pp.198-201
• /
• 1999

An automated storage and retrieval machinery for high rack warehouse systems is developed in order to stack the various kinds of productions. However, according to increase in the rack height, the long lead time should be taken. In stacker crane systems, the variations of the lifting height and the load generate the vibration of lifting machine, and it makes a Position control to be difficult Therefore, the reduction of vibration will be important factor for saving the lead time and the damage of Productions. This paper deals with a position control of stacker crane in automated high rack warehouse system by using a gain-scheduled control algorithm via a LMI method, where the variations of elastic coefficient of the stacker crane's post are considered.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.3
• /
• pp.1188-1203
• /
• 2020

The human expression contains a lot of information that can be used to detect complex conditions such as pain and fatigue. After deep learning became the mainstream method, the traditional feature extraction method no longer has advantages. However, in order to achieve higher accuracy, researchers continue to stack the number of layers of the neural network, which makes the real-time performance of the model weak. Therefore, this paper proposed an expression recognition framework based on densely concatenated convolutional neural networks to balance accuracy and latency and apply it to humanoid robots. The techniques of feature reuse and parameter compression in the framework improved the learning ability of the model and greatly reduced the parameters. Experiments showed that the proposed model can reduce tens of times the parameters at the expense of little accuracy.

• Annual Conference of KIPS
• /
• 2016.10a
• /
• pp.66-67
• /
• 2016

최근 수많은 국 내외 글로벌 기업들이 클라우드 자원의 제공자 겸 소비자 역할을 하는 프라이빗 IaaS 클라우드 환경을 구축하고 있는 추세이며 이를 위해 오픈소스 클라우드 플랫폼인 오픈스택(OpenStack)이 많이 사용되고 있다. 이 논문에서는 대규모 빅데이터 처리를 위해 오픈스택 클라우드 환경의 가상머신 라이브 마이그레이션 기법을 사용할 경우 발생할 수 있는 문제점을 분석한다. 이러한 문제점에 대하여 가상머신에서 빅데이터 연산 처리 시 스토리지 병목현상을 해결하기 위한 마이그레이션 기법을 제시한다.

• Annual Conference of KIPS
• /
• 2002.11b
• /
• pp.1387-1390
• /
• 2002

자바가상머신은 플랫폼에 독립적인 실행을 위해서 바이트코드라고 하는 스택(stack) 기반의 가상 기계어를 사용하므로 실행 속도가 느리다는 단점이 있다. 특히 루프문을 포함하는 자바프로그램을 자바가상머신에서 수행 시키면 루프에 관련된 몇 개의 동일한 바이트코드가 루프의 실행 횟수만큼 반복적으로 인터프리트해서 수행하므로 상당한 성능 저하를 유발한다. 본 논문에서는 이런 비효율적인 성능상의 문제점을 개선하기 위해 루프를 수행하는 새로운 바이트코드를 정의 및 구현하고, 이를 실제 클래스 파일에 적용하기 위한 코드 변경 절차와 방법을 제시한다. 제안된 바이트코드를 사용해서 루프의 처리 속도를 개선할 경우, 클래스 파일의 크기를 줄일 수 있을 뿐만 아니라 간단한 성능 평가를 통해서 자바가상머신의 성능 개선 효과를 확인할 수 있다.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.3 s.96
• /
• pp.91-102
• /
• 1999

Fieldbus provides real-time data communication among field devices in the manufacturing automation and process control systems. In this study, an experimental model of fieldbus-based manufacturing automation system is developed. Experimental model consists of two robots, two conveyor belts, NC machine, PLC, sensors and operator station. These machines are interconnected into the Profibus network, and exchange their data through the services provided by FMS(Fieldbus Message Specification), which is the application layer protocol of Profibus. The experimental model is used to measure the network-induced delay of variable and file data transmitted through FMS services. Network-induced delays are collected and analyzed on each sublayer of Profibus protocol stack. The results obtained from the experiment of this study can be effectively utilized when fieldbus is implemented on the practical manufacturing automation systems.