• 대한전자공학회논문지SD
• /
• 제42권2호
• /
• pp.65-72
• /
• 2005

공정기술과 EDA 툴의 발전에 따라서 하나의 실리콘 다이(Die)에 많은 IP가 집적되고 멀티프로세서가 포함되는 SoC 구조가 가능해지고 있다 그러나 대부분의 기존 SoC 버스는 공유버스 구조라는 문제점으로 인해 통신의 병목현상이 발생하고 이는 전체 시스템 성능을 저하시키는 요인이 된다. 많은 경우에 멀티프로세서 시스템의 성능은 CPU 자체의 속도보다는 효율적인 통신과 균형있는 연산의 분배가 좌우하게 된다 따라서 충분한 SoC 버스 대역폭(Bandwidth)을 확보하기 위한 하나의 해결책으로 크로스바 라우터(Crossbar Router)를 이용하여 효율적인 온 칩 버스구조인 SoC Network Architecture(SNA)를 제안한다. 제안된 SNA구조는 다중 마스터(multi-master)에 대해 다중 채널(multi-channel)을 제공하여 통신의 병목현상을 크게 줄일 수 있으며 뛰어난 확장성을 지원한다. 제안된 구조에 따라 모델 시스템을 설계하고 시뮬레이션을 진행한 결과 AMBA AHB 버스에 비해 평균 $40\%$ 이상 효율이 증가했다.

• 비파괴검사학회지
• /
• 제30권4호
• /
• pp.338-344
• /
• 2010

웨이브가이드 초음파센서를 적용하여 소듐냉각 고속로 KALIMER-600의 원자로 노심과 내부구조물의 소듐내부가시화를 수행하는 이중회전 구동 C-스캔 시스템과 소프트웨어 프로그램 Under-Sodium MultiVIEW를 개발하였다. 이중회전 구동 C-스캔 시스템은 KALIMER-600 원자로 헤드의 이중회전 플러그를 모사하여 설계 제작하였으며 웨이브가이드 초음파센서에 초음파 펄스를 송수신할 수 있는 고출력 초음파 시스템과 스캐너 구동 제어 장치를 구축하였다. Under-Sodium MultiVIEW 프로그램은 이중회전 스캐너의 구동을 제어하면서 웨이브가이드 초음파센서에 초음파 신호를 송수신하여 영상 처리를 수행하는 소듐내부가시화 프로그램으로서 LabVIEW 그래픽 프로그램 언어를 기반으로 개발되었다. 이중회전 구동에 의한 수중 C-스캔 성능시험을 수행하여 Under-Sodium MultiVIEW 프로그램의 가시화 성능을 실험적으로 검증하였다.

• 한국초전도ㆍ저온공학회논문지
• /
• 제24권3호
• /
• pp.52-56
• /
• 2022

Single-grain (c-normal or c-parallel) and multi-grain YBCO superconductors were prepared by a melt growth process with/without seeding. The magnetic levitation force and trapped magnetic field at liquid N₂ temperature (77 K) of the YBCO superconductors were investigated. Samples for the levitation force measurement were zero-field cooled (ZFC) to 77 K, and samples for trapped field measurement were field-cooled (FC) using Nd magnets. As for the magnetic levitation force, the c-normal, single grain sample showed the largest value, whereas the multi-grain sample showed the lowest value. The trapped magnetic field of the c-normal and c-parallel single-grain samples was 4-5 times that of the multi-grain sample. In addition, as the external magnetic field (the number of magnets) increased, the both properties increased proportionally. These results were explained in terms of the orientation dependence of the levitation forces and the magnetic field trapping capability of the YBCO superconductor.

• 한국컴퓨터정보학회논문지
• /
• 제21권4호
• /
• pp.1-8
• /
• 2016

In small embedded systems including IoT devices, memory size is very small and it is important to reduce memory amount for execution of application programs. For multi-threaded applications, stack may consume a large amount of memory because each thread has its own stack of sufficiently large size for worst case. This paper presents an implementation of single stack multi-threading, called SSThread (Single Stack Thread), by sharing a stack for all threads to reduce stack memory size. By using SSThread, multi-threaded applications can be programmed based on normal C language environment and there is no requirement of transporting multi-threading operating systems. It consists of several library functions and various C macro definitions. Even though some functional restrictions in comparison to operating systems supporting complete multi-thread functionalities, it is very useful for small embedded systems with tiny memory size and it is simple to setup programming environment for multi-thread applications.

• 설비공학논문집
• /
• 제23권2호
• /
• pp.126-131
• /
• 2011

Cyclone, a type of particle collector widely used in the field of ambient sampling and industrial particulate control, is the principal type of gas-solids separator that uses a centrifugal force. The goal of this study is to design and evaluate the cyclone that can be used for the household vacuum cleaners. A multi cyclone with a 1st cyclone and several 2nd cyclones is designed to improve dust collection efficiency. The dust collection efficiency and the suction power of 1st cyclone are evaluated. And the dust collection efficiency and the suction power of multi cyclone are evaluated according to various sizes of inlet and vortex finder. As a result, a cone shape porous filter has better dust collection efficiency than a cylinder shape porous filter. The dust collection efficiency of a multi cyclone is 3.5% greater than that of a single cyclone.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2014년도 제46회 동계 정기학술대회 초록집
• /
• pp.322.1-322.1
• /
• 2014

Many research groups have studied tandem or multi-junction cells to overcome this low efficiency and degradation. In multi-junction cells, band-gap engineering of each absorb layer is needed to absorb the light at various wavelengths efficiently. Various absorption layers can be formed using multi-junctions, such as hydrogenated amorphous silicon carbide (a-SiC:H), amorphous silicon germanium (a-SiGe:H) and microcrystalline silicon (${\mu}c$-Si:H), etc. Among them, ${\mu}c$-Si:H is the bottom absorber material because it has a low band-gap and does not exhibit light-induced degradation like amorphous silicon. Nevertheless, ${\mu}c$-Si:H requires a much thicker material (>2 mm) to absorb sufficient light due to its smaller light absorption coefficient, highlighting the need for a high growth rate for productivity. ${\mu}c$-SiGe:H has a much higher absorption coefficient than ${\mu}c$-Si:H at the low energy wavelength, meaning that the thickness of the absorption layer can be decreased to less than half that of ${\mu}c$-Si:H. ${\mu}c$-SiGe:H films were prepared using 40 MHz very high frequency PECVD method at 1 Torr. SiH4 and GeH4 were used as a reactive gas and H2 was used as a dilution gas. In this study, the ${\mu}c$-SiGe:H layer for triple solar cells applications was performed to optimize the film properties.

• International Journal of Internet, Broadcasting and Communication
• /
• 제16권3호
• /
• pp.192-198
• /
• 2024

This paper investigates the application of multi-agent deep reinforcement learning in the fighting game Samurai Shodown using Proximal Policy Optimization (PPO) and Advantage Actor-Critic (A2C) algorithms. Initially, agents are trained separately for 200,000 timesteps using Convolutional Neural Network (CNN) and Multi-Layer Perceptron (MLP) with LSTM networks. PPO demonstrates superior performance early on with stable policy updates, while A2C shows better adaptation and higher rewards over extended training periods, culminating in A2C outperforming PPO after 1,000,000 timesteps. These findings highlight PPO's effectiveness for short-term training and A2C's advantages in long-term learning scenarios, emphasizing the importance of algorithm selection based on training duration and task complexity. The code can be found in this link https://github.com/Lexer04/Samurai-Shodown-with-Reinforcement-Learning-PPO.

• E2M - 전기 전자와 첨단 소재
• /
• 제9권6호
• /
• pp.600-604
• /
• 1996

High-T$_{c}$, superconducting wire is very important element for the application of electrical power systems. But it is very difficult to develope the long high T, wire with excellent properties. BiSrCaCuO multi-layer tapes are fabricated by a rolling method and pressing method sintered for several step at 840.deg. C. The critical current densities of 637 filament multi-layer tapes sintering 100 hours fabricated by the rolling method and pressing method are 1.3*10$^{4}$ A/cm$^{2}$ and 5.5*10$^{3}$ A/cm$^{2}$. The critical cur-rent densites of multi-layer tapes made by rolling method are found to be better than those fabricated by the powder-in-tube method and pressing process. As result, the rolling method is the best way to fabricated the multi-layer filament.t.

• IEIE Transactions on Smart Processing and Computing
• /
• 제5권1호
• /
• pp.55-62
• /
• 2016

In this paper, we propose a hybrid multi.system-on-chip (H-MSoC) architecture that provides a high-flexibility system in a rapid development time. The H-MSoC approach provides a flexible system-on-chip (SoC) architecture that is easy to configure for physical- and application-layer development. The physical- and application-layer aspects are dynamically designed and modified; hence, it is important to consider a design methodology that supports rapid SoC development. Physical layer development refers to intellectual property cores or other modular hardware (HW) development, while application layer development refers to user interface or application software (SW) development. H-MSoC is built from multi-SoC architectures in which each SoC is localized and specified based on its development focus, either physical or application (hybrid). Physical HW development SoC is referred to as physical-SoC (Phy-SoC) and application SW development SoC is referred to as application-SoC (App-SoC). Phy-SoC and App-SoC are connected to each other via Ethernet. Ethernet was chosen because of its flexibility, high speed, and easy configuration. For prototyping, we used a LEON3 SoC as the Phy-SoC and a ZYNQ-7000 SoC as the App-SoC. The proposed design was proven in real-time tests and achieved good performance.

• International Journal of Internet, Broadcasting and Communication
• /
• 제16권3호
• /
• pp.205-211
• /
• 2024

Effective warehouse management requires advanced resource planning to optimize profits and space. Robots offer a promising solution, but their effectiveness relies on embedded artificial intelligence. Multi-agent reinforcement learning (MARL) enhances robot intelligence in these environments. This study explores various MARL algorithms using the Multi-Robot Warehouse Environment (RWARE) to determine their suitability for warehouse resource planning. Our findings show that cooperative MARL is essential for effective warehouse management. IA2C outperforms MAA2C and VDA2C on smaller maps, while VDA2C excels on larger maps. IA2C's decentralized approach, focusing on cooperation over collaboration, allows for higher reward collection in smaller environments. However, as map size increases, reward collection decreases due to the need for extensive exploration. This study highlights the importance of selecting the appropriate MARL algorithm based on the specific warehouse environment's requirements and scale.