• Journal of Information Processing Systems
• /
• 제16권6호
• /
• pp.1359-1371
• /
• 2020

In transmitting and receiving such a large amount of data, reliable data communication is crucial for normal operation of a device and to prevent abnormal operations caused by errors. Therefore, in this paper, it is assumed that an error correction code (ECC) that can detect and correct errors by itself is used in an environment where massive data is sequentially received. Because an embedded system has limited resources, such as a low-performance processor or a small memory, it requires efficient operation of applications. In this paper, we propose using an accelerated ECC-decoding technique with a graphics processing unit (GPU) built into the embedded system when receiving a large amount of data. In the matrix-vector multiplication that forms the Hamming code used as a function of the ECC operation, the matrix is expressed in compressed sparse row (CSR) format, and a sparse matrix-vector product is used. The multiplication operation is performed in the kernel of the GPU, and we also accelerate the Hamming code computation so that the ECC operation can be performed in parallel. The proposed technique is implemented with CUDA on a GPU-embedded target board, NVIDIA Jetson TX2, and compared with execution time of the CPU.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2004년도 추계학술대회 논문집
• /
• pp.720-723
• /
• 2004

Lever mechanisms are usually employed to enlarge output displacement in precision stages. In this study, theoretical analysis of a lever is presented including bending effect and relation between dimension parameters and an objective function. The objective function is chosen as multiplication of magnification ratio and forcedisplacement transmission. Through theoretical analysis, this study presents optimal values for the parameters and the analysis is verified by finite element method.

• 전자공학회논문지B
• /
• 제32B권1호
• /
• pp.180-188
• /
• 1995

A modified Kohonen's self-organizing feature map (SOFM) algorithm which has binary reinforcement function and a constant learning rate is proposed. In contrast to the time-varing adaptaion gain of the original Kohonen's SOFM algorithm, the proposed algorithm uses a constant adaptation gain, and adds a binary reinforcement function in order to compensate for the lowered learning ability of SOFM due to the constant learning rate. Since the proposed algorithm does not have the complicated multiplication, it's digital hardware implementation is much easier than that of the original SOFM.

• 충청수학회지
• /
• 제33권4호
• /
• pp.497-504
• /
• 2020

A weight ω on the positive half real line [0, ∞) is a positive continuous function such that ω(s + t) ≤ ω(s)ω(t), for all s, t ∈ [0, ∞), and ω(0) = 1. The weighted convolution Banach algebra L1(ω) is the algebra of all equivalence classes of Lebesgue measurable functions f such that ‖f‖ = ∫0∞|f(t)|ω(t)dt < ∞, under pointwise addition, scalar multiplication of functions, and the convolution product (f ⁎ g)(t) = ∫0t f(t - s)g(s)ds. We give a sufficient condition on a weight function ω(t) in order that L1(ω) has a bounded approximate identity.

• 대한전자공학회논문지SD
• /
• 제40권4호
• /
• pp.51-57
• /
• 2003

통신망 및 그 이외의 네트워크 환경의 발전은 사회적으로 중요한 문제를 발생시켰다. 이러한 문제점 중 가장 중요한 것이 네트워크 보안 문제이다. 보안과 관련된 문제점들은 해킹, 크랙킹과 같은 방법으로 반 보안 분야를 확장시키며 발전되었다. 새로운 암호 알고리즘의 발달 없이 해커나 크래커로부터 데이터를 보호하기 위해서는 기존과 같이 키의 길이를 증대하거나 처리 데이터의 양을 증대시키는 방법 밖에는 없다. 본 논문에서는 공개키 암호 알고리즘의 몽고메리 승산부에서 처리속도를 감소시키기 위한 M3 알고리즘을 제안하였다. 매트릭스 함수 M(·)과 룩업테이블을 사용하는 제안된 M3 알고리즘은 몽고메리 승산부의 반복 연산부를 선택적으로 수행하게 된다. 이러한 결과로 변형된 반복 변환 부분은 기존 몽고메리 승산기에 비하여 30%의 처리율 향상을 가져왔다. 제안된 몽고메리 승산 M3 알고리즘은 캐리 생성부의 어레이 배열과 가변 길이 오퍼랜드 감소로 인한 병목 현상을 줄일 수 있다. 그러므로 본 논문에서는 제안된 M3 알고리즘을 공개키 암호시스템의 대표적인 시스템인 RSA에 적용하여 M3-RSA를 설계하였으며 설계 및 모의실험은 Synopsys ver 1999.10을 사용하였다. M3 알고리즘은 기존 승산알고리즘에 비하여 30%의 처리속도 증가를 보임으로서 크랙 및 처리율 향상에 영향이 많은 공개키 암호시스템에 적합하리라 사료된다.

• 전자공학회논문지B
• /
• 제30B권10호
• /
• pp.69-76
• /
• 1993

In this paper we propose a design of a digital neuron processor using the residue number system for efficient matrix.vector multiplication involved in neural processing. Since the residue number system needs no carry propagation for modulus operations, the neuron processor can perform multiplication considerably fast. We also propose a high speed algorithm for computing the sigmoid function using the specially designed look-up table. Our method can be implemented area-effectively using the current technology of digital VLSI and siumlation results positively demonstrate the feasibility of our method. The proposed method would expected to adopt for application field of digital neural network, because it could be realized to currently developed digital VLSI Technology.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1987년도 전기.전자공학 학술대회 논문집(II)
• /
• pp.1195-1198
• /
• 1987

This paper develops algorithms of element generation, addition, multiplication and inversion based on GF($2^m$). Since these algorithms are implemented by general purpose computer, these are more efficient than the conventional algorithms(Table Lookup, Euclid's Algorithm) in each operation. It is also implied that they can be applied to not only the normally defined elements but the arbitrarily defined ones for constructing multi-valued logic function.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 1991년도 추계학술대회 논문집
• /
• pp.1-4
• /
• 1991

EEPROM technology has been used for storing analog weights as charge in a nitride layer between gate and channel of a field effect transistor. In the view of integrity and fabrication process, it is essentially required that SONOSFET is capable of performing synapse function as a basic element in an artificial neural networks. This work has introduced the VLSI implementation for synapses including current study and also investigated physical characteristics to implement synapse circuit using SONOSFET memories. Simulation results are shown in this work. It is proposed that multiplication of synapse element using SONOSFET memories will be developed more compact implementation under Present fabrication processes.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2002년도 추계학술대회 논문집 전기기기 및 에너지변환시스템부문
• /
• pp.21-23
• /
• 2002

We designed and manufactured helical magnetic flux compression generator, in which, the current density was reduced by increasing the number of wires by stages, and the voltage between wires was reduced by decreasing the time rate of inductance change. The figure of merit and the energy multiplication ratio of the generator were measured as a function of current density flowing through coil and their characteristics were analyzed. When the current density of coil was more than 250 kA/cm, the figure of merit and the energy multiplication ratio were decreased rapidly.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 1996년도 추계학술대회 논문집
• /
• pp.1-4
• /
• 1996

EEPROM technology has been used for storing analog weights as charge in a nitride layer between gate and channel of a field effect transistor. In the view of integrity and fabrication process, it is essentially required that SONOSFET is capable of performing synapse function as a basic element in an artificial neural networks. This work has introduced the VLSI implementation for synapses including current study and also investigated physical characteristics to implement synapse circuit using SONOSFET memories. Simulation results are shown in this work. It is proposed that multiplication of synapse element using SONOSFET memories will be developed more compact implementation under Present fabrication processes.