• Annual Conference of KIPS
• /
• 2015.10a
• /
• pp.253-254
• /
• 2015

계산과학분야에서 컴퓨팅자원을 사용하는 사용자들은 수천 개의 CPU 규모의 클러스터단위로 컴퓨팅 자원을 사용한다. 자원의 크기에 따라 작업 실행 시간이 줄어들기 때문에 사용자들이 정확하고 빠른 연구결과를 얻기 위해서는 많은 컴퓨팅자원이 필요하다. 하지만 컴퓨팅자원의 한계와 비용의 문제로 모든 사용자들이 원하는 자원을 할당 받지 못한다. 본 논문에서는 컴퓨팅자원을 가상머신 클러스터 단위로 제공하는 방법과 자원의 낭비를 줄이기 위한 가상머신 동적 할당방법을 구현하였다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.4
• /
• pp.717-724
• /
• 2020

Due to the important role played by emotion in human interaction, affective computing is dedicated in trying to understand and regulate emotion through human-aware artificial intelligence. By understanding, emotion mental diseases such as depression, autism, attention deficit hyperactivity disorder, and game addiction will be better managed as they are all associated with emotion. Various studies for emotion recognition have been conducted to solve these problems. In applying machine learning for the emotion recognition, the efforts to reduce the complexity of the algorithm and improve the accuracy are required. In this paper, we investigate emotion Electroencephalogram (EEG) feature reduction and classification using Stack AutoEncoder (SAE) and Long-Short-Term-Memory/Recurrent Neural Networks (LSTM/RNN) classification respectively. The proposed method reduced the complexity of the model and significantly enhance the performance of the classifiers.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.64 no.1
• /
• pp.7-13
• /
• 2015

This paper presents mathematical models for high speed permanent-magnet synchronous machine. The mathematical method with two successive steps is used to estimate design parameter as well as the output power. At first, mathematical model for a linkage flux problem is employed to calculate the number of winding turns and stack length of armature core. The magnetic circuit model for an induced voltage and the electric circuit model for a current are modeled. The output powers of the electrical generator were evaluated by the mathematical techniques. The results of this mathematical methods predict the specifications of the machine and can be applied in the design stage of the electrical machine.

• International Journal of Computer Science & Network Security
• /
• v.24 no.10
• /
• pp.182-190
• /
• 2024

The use of Short Message Services (SMS) as a mechanism of communication has resulted to loss of sensitive information such as credit card details, medical information and bank account details (user name and password). Several Machine learning-based approaches have been proposed to address this problem, but they are still unable to detect modified SMS spam messages more accurately. Thus, in this research, a stack- ensemble of four machine learning algorithms consisting of Random Forest (RF), Logistic Regression (LR), Multilayer Perceptron (MLP), and Support Vector Machine (SVM), were employed to detect more accurately SMS spams. The simulation was carried out using Python Scikit- learn tools. The performance evaluation of the proposed model was carried out by benchmarking it with an existing model. The evaluation results showed that the proposed model has an increase of 3.03% of accuracy, 8.94% of Recall, 2.17% of F-measure; and a decrease of 4.55% of Precision over the existing model. In conclusion, the ensemble method performed better than any individual algorithms and can be adopted by the Network service providers for better Quality of Service.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.05d
• /
• pp.1001-1006
• /
• 2002

KVM에서의 검증단계는 크게 두가지로 나된다. 일반적으로 처리되는 Off-device에서의 사전검증과 실제 VM에 탑재되어 수행되는 in-device에서의 검증이다. 전자의 경우, 실행시 검증을 쉽게 하기 위해서 클래스 파일에 특별한 속성을 가진 스택맵을 추가한다. 이렇게 추가된 스택맵은 KVM상에서 보다 효율적인 검증을 위한 것으로써, 본 논문에서는 이러한 스택맵을 분석하였다.

• Journal of KIISE:Software and Applications
• /
• v.32 no.3
• /
• pp.215-226
• /
• 2005

Loop invariant code motion (LICM) optimization includes relatively heavy code analyses, thus being not readily applicable to Java Just-In-Time (JIT) compilation where the JIT compilation time is part of the whole running time. 'Classical' LICM optimization first analyzes the code and constructs both the def-use chains and the use-def chains. which are then used for performing code motions. This paper proposes a light-weight LICM algorithm, which requires only the def-use chains of loop invariant code (without use-def chains) by exploiting the fact that the Java virtual machine is based on a stack machine, hence generating code with simpler patterns. We also propose two techniques that allow more code motions than classical LICM techniques. First, unlike previous JIT techniques that uses LICM only in single-path loops for simplicity, we apply LICM to multi-path loops (natural loops) safely for partially redundant code. Secondly, we move loop-invariant, partially-redundant null pointer check code via predication support in Itanium. The proposed techniques were implemented in a JIT compiler for Itanium processor on ORP (Open Runtime Platform) Java virtual machine of Intel. On SPECjvrn98 benchmarks, the proposed technique increases the JIT compilation overhead by the geometric mean of 1.3％, yet it improves the total running time by the geometric mean of 2.2％.

• Proceedings of the KIPE Conference
• /
• 2008.06a
• /
• pp.624-627
• /
• 2008

It is generally used the boost converter for the welding machine using fuel cell. But this type generates the duplicate loss between boost converter and inverter. As a result, the whole efficiency is lower than conventional inverter arc welder and the cost is expensive due to additional switching device. Therefore, we proposed the inverter rectification type DC-DC converter with boost converter function in fuel cell based TIG welder. The performance of the proposed technique is evaluated on a 1.2kW fuel cell stack based experimental prototype circuit.

• Journal of Advanced Marine Engineering and Technology
• /
• v.40 no.7
• /
• pp.629-634
• /
• 2016

This paper describes the implementation of a ZigBee sensor node that can be utilized as a multiband and machine to machine (M2M) communication gateway. The IEEE 802.15.4-2003 standard was used as the wireless network frequency band. Ember's Type EM357 SoC was used as the transmission and reception device to perform the communication function, and it was also used for both the main M2M gateway and the sensor node. For the implementation of the operating protocol, EmberZNet Stack 4.5.4 from the Ember Corporation was used. The measurement of the reception sensitivity in the receiving module and the actual output signal from the reference were obtained from the transmission of a packet, and the packet included the M2M gateway within the attached ZigBee sensor. The packet error rate was measured as 0% with a -98 dBm reception sensitivity at the ZigBee frequency. In addition, excellent current characteristics of the ZigBee modules were shown by the implementation of the low-power circuit.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.11
• /
• pp.4246-4267
• /
• 2020

As the network goes deep into all aspects of people's lives, the number and the complexity of network traffic is increasing, and traffic classification becomes more and more important. How to classify them effectively is an important prerequisite for network management and planning, and ensuring network security. With the continuous development of deep learning, more and more traffic classification begins to use it as the main method, which achieves better results than traditional classification methods. In this paper, we provide a comprehensive review of network traffic classification based on deep learning. Firstly, we introduce the research background and progress of network traffic classification. Then, we summarize and compare traffic classification based on deep learning such as stack autoencoder, one-dimensional convolution neural network, two-dimensional convolution neural network, three-dimensional convolution neural network, long short-term memory network and Deep Belief Networks. In addition, we compare traffic classification based on deep learning with other methods such as based on port number, deep packets detection and machine learning. Finally, the future research directions of network traffic classification based on deep learning are prospected.

• KIISE Transactions on Computing Practices
• /
• v.24 no.3
• /
• pp.129-137
• /
• 2018

NVMe(Non-Volatile Memory Express) SSD(Solid State Drive) is a high-performance storage that makes use of flash memory as a storage cell, PCIe as an interface and NVMe as a protocol on the interface. It supports multiple I/O queues which makes it feasible to process parallel-I/Os on multi-core environments and to provide higher bandwidth than SATA SSDs. Hence, NVMe SSD is considered as a next generation-storage for data-center and cloud computing system. However, in the virtualization system, the performance of NVMe SSD is not fully utilized due to the bottleneck of the software I/O stack. Especially, when it uses I/O stack of the hypervisor or the host operating system like Xen and KVM, I/O performance degrades seriously due to doubled-I/O stack between host and virtual machine. In this paper, we propose a new I/O engine, called Direct-AIO (Direct-Asynchronous I/O) engine, that can access NVMe SSD directly for I/O performance improvements on QEMU emulator. We develop our proposed I/O engine and analyze I/O performance differences between the existed I/O engine and Direct-AIO engine.