• 대한전자공학회논문지
• /
• 제22권6호
• /
• pp.114-123
• /
• 1985

이 논문은 packet switching 방식에 의한 공중 computer 통신망 개발에 관한 4편의 논문중 제 2부의 논문으로 제 1부의 KORNET의 개요 및 netwo.k managementcenter (NMC)개발에관한 논문에 이어 KO-RNET의 설계와 networhnode Processor(NNP)의 개발에 대해 기술한다. KORNET은 3개의 NNP와 하나의 NMC 로 일차 구성하였는데, NNP는 MC68000 microprocessor를 이용한 multiprocessor system으로 구현되었고, HMC는 중형 computer인 Mv/8000 system을 사용하여 개발하였다. KORNET에서의 packet service 방식은 virtual circuit(VC) 방식으로 하고 routing은 node나 선로의 상태변화에 쉽게 대처할 수 있는 분산적응방식(distributed adaptive routinB)을 사용하였다. 또한, buffo. management는 dy-namic sharing 방식을 채택하여 storage의 사용에 대한 효율성을 높였다. NNP system의 hardware는 modularity를 고려하여 확장이 용이하게 하였으며, software는 CCITT 권고사항 X.25, X.3, X.28, X.29등을 따라 구현하였다.

• 한국통신학회논문지
• /
• 제21권12호
• /
• pp.3144-3153
• /
• 1996

ATM switching system is made up of transport network and control newrk according to its functions. The control device, basic part of control network must be developed before developing any other functions, and control device must be stable and need high reliability. Out distributed ATM switching system consists of several ALSs that provides variable local call services, and an ACS that interconnect among several ALSs. Eech ALS has CCCP that takes charage of call and connection control functions, and ACS has an OMP that takes charge of OA&M(Operation, Administration and Maintenance) functios. In this paper, we analyzed the performance evaluation of control device that manipulate subscriber's call based on ITU-T Q.2931 standard protocol messages and Interprocessor communication messages. As a result of simulation when the number of ALS is under 22, as the call arrival rate increase the processor utilization of CCCP increase rapidly than that of OMP. When the number of ALS is incremented to 22, the processor utilization of CCCP is balanced with the of OMP, and when the number of ALS exceeds 22, the processor utiliztion of OMP increase rapidly. Also if messary processing time of OMP is 1.35 times that of CCCP, processor utilizations of CCCP and OMP is equal.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2011년도 춘계학술발표대회
• /
• pp.46-49
• /
• 2011

This paper is to implement MoIP platform to send and receive video and audio at the same time by using high-performance Dual Core Processor. Even if Wall-Pad key component of a home network system is released by using embedded processors, it's lacking of performance in terms of multimedia processing and feature of video telephony through which video and voice are exchanged simultaneously. The main reason could be that embedded processors currently being used do not provide enough performance to support both MoIP call features and various home network features simultaneously. In order to solve these problems, Dual processor could be used, but in the other hands it brings another disadvantage of high cost. Therefore, this study is to solve the home automation features and video telephony features by using Dual Core Processor based on ARM 11 Processor and implement the MoIP Wall-Pad which can reduce the board design costs and component costs, and improve performance. The platform designed and implemented in this paper verified performance of MoIP to exchange the video and voice at the same time under the situation of Ethernet network.

• 한국지능시스템학회논문지
• /
• 제15권2호
• /
• pp.224-229
• /
• 2005

일반적으로 유전 알고리즘은 전형적인 프로세서에서 수행할 경우 매우 큰 시간 공간 복잡도를 가진다. 따라서 유전 알고리즘 처리를 위해서는 고성능$\cdot$고가격의 프로세서를 필요로 하게 된다. 또한 이것은 유전 알고리즘을 소형 이동 로봇과 같이 비교적 간단한 룰을 필요로 하는 실제 하드웨어에 적용하는데 있어 큰 장벽으로 작용한다. 이러한 문제의 해결을 위해, 본 논문에서는 유전 알고리즘의 신속한 처리를 위해 강화된 프로세서 구조를 보인다. 정렬 네트워크와 residue number system (RNS)를 이용하여 일반적인 프로세서의 구조를 유전 알고리즘의 처리에 효율적이도록 강화할 수 있다. 정렬 네트워크는 유전 알고리즘에 필수적인 해들의 품질 비교를 하드웨어적으로 처리할 수 있게 하여 수행에 요구되는 시간을 줄일 수 있다. RNS는 산술 연산의 속도를 좌우하는 bit 사이즈를 줄여 전체적인 로직의 사이즈를 줄이고, 산술 연산의 처리 속도를 빠르게 할 수 있다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -1
• /
• pp.313-316
• /
• 2002

This paper presents the design and implementation of a crypto processor, a special-purpose microprocessor optimized for the execution of cryptography algorithms. This crypto processor can be used fur various security applications such as storage devices, embedded systems, network routers, etc. The crypto processor consists of a 32-bit RISC processor block and a coprocessor block dedicated to the SEED and triple-DES (data encryption standard) symmetric key crypto (cryptography) algorithms. The crypto processor has been designed and fabricated as a single VLSI chip using 0.5 $\mu\textrm{m}$ CMOS technology. To test and demonstrate the capabilities of this chip, a custom board providing real-time data security for a data storage device has been developed. Testing results show that the crypto processor operates correctly at a working frequency of 30MHz and a bandwidth o1240Mbps.

• International journal of advanced smart convergence
• /
• 제11권1호
• /
• pp.28-35
• /
• 2022

This paper presents a performance evaluation of real-time Linux for industrial real-time platforms. On industrial platforms, multicore processors are popular due to their work distribution efficiency and cost-effectiveness. Multicore processors, however, are not designed for applications with real-time constraints, and their performance capabilities depend on their core configurations. In order to assess the feasibility of a multicore processor for real-time applications, we conduct a performance evaluation of a general processor and a low-power processor to provide an experimental environment of real-time Linux on both Xenomai and RT-preempt considering the multicore configuration. The real-time performance is evaluated through scheduling latency and in an environment with loads on the CPU, memory, and network to consider an actual situation. The results show a difference between a low-power and a general-purpose processor, but from developer's point of view, it shows that the low-power processor is a proper solution to accommodate low power situations.

• 전자공학회논문지A
• /
• 제33A권2호
• /
• pp.1-7
• /
• 1996

In this paper, we analyzed the packet blocking probability of a finite buffer in a network interface unit. In general, a network interface unit which provides a means of interface between the network and computer has a microprocessor and a protocol processor for the network access protocols. It also has a receive buffer for the arriving packets from the network which is served by the microprocessor with service interruption by the protocol processor. In this paper, we modeled the receive buffer as a discrete time server with service interruption, and obtained the packet blocking probability using the mini-slot approximation.

• 전기전자학회논문지
• /
• 제21권1호
• /
• pp.92-95
• /
• 2017

본 논문에서는 적은 면적의 GP-GPU에서 성능을 향상시키기 위한 방법을 제안한다. 본 논문에서는 superscalar와 같이 과도하게 스케줄링 복잡성을 증가시키지 않는 대신 단순한 코어의 수를 늘려 성능을 극대화 시키는 방법을 제안한다. GP-GPU를 구성하는 Stream Processor의 구조를 단순화한다. 또한, Warp Schedule에서 thread 할당을 어플리케이션에 적합한 방법을 개발하여 성능을 개선한다. 성능을 검증하는 방안으로 neural network의 한 분야인 딥러닝에 대한 스레드 할당방식을 제안한다. Neural Network 알고리즘의 경우 Intel CPU 대비 90%에서 ARM Cortex-A15 4 core 대비 98% 성능 향상을 확인할 수 있었다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 추계학술대회 논문집
• /
• pp.738-742
• /
• 1996

This paper presents the automatic monitoring system for the small and medium sized manufacturing system. The monitoring system was composed of main controller, network card and monitoring sub controller for the unit machine. PC was used for the main controller and monitoring controller, which has the same hardware with the network card, was developed using Intel 80196 microprocessor.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2008년도 추계학술발표논문집
• /
• pp.191-196
• /
• 2008

The intent of this paper is to describe a neural network structure called dynamic neural processor(DNP), and examine how it can be used in developing a learning scheme for computing robot inverse kinematic transformations. The architecture and learning algorithm of the proposed dynamic neural network structure, the DNP, are described. Computer simulations are provided to demonstrate the effectiveness of the proposed learning using the DNP.