• Geophysics and Geophysical Exploration
• /
• v.19 no.3
• /
• pp.145-152
• /
• 2016

We used graphics processing units for an efficient time-domain 3D wave propagation modeling. Since graphics processing units are designed for massively parallel processes, we need to optimize the calculation and memory management to fully exploit graphics processing units. We focused on the memory management and examined the performance of programs with respect to the memory management methods. We also tested the effects of memory transfer on the performance of the program by varying the order of finite difference equation and the size of velocity models. The results show that the memory transfer takes a larger portion of the running time than that of the finite difference calculation in programs transferring whole 3D wavefield.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.837-852
• /
• 2021

Recurrent neural network (RNN) architectures have been used for language modeling (LM) tasks that require learning long-range word or character sequences. However, the RNN architecture is still suffered from unstable gradients on long-range sequences. To address the issue of long-range sequences, an attention mechanism has been used, showing state-of-the-art (SOTA) performance in all LM tasks. A differentiable neural computer (DNC) is a deep learning architecture using an attention mechanism. The DNC architecture is a neural network augmented with a content-addressable external memory. However, in the write operation, some information unrelated to the input word remains in memory. Moreover, DNCs have been found to perform poorly with low numbers of weight parameters. Therefore, we propose a robust memory deallocation method using a limited retention vector. The limited retention vector determines whether the network increases or decreases its usage of information in external memory according to a threshold. We experimentally evaluate the robustness of a DNC implementing the proposed approach according to the size of the controller and external memory on the enwik8 LM task. When we decreased the number of weight parameters by 32.47%, the proposed DNC showed a low bits-per-character (BPC) degradation of 4.30%, demonstrating the effectiveness of our approach in language modeling tasks.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.9C
• /
• pp.880-886
• /
• 2009

Recently, the memory is used in many electronic devices, thus, the many researchers make a study of the memory. To increase a storage capacity per memory block, the researchers study for reducing the fabrication process of memory and multi-level cell memory which is storing more than 2-bits in a cell. However, the multi-level cell memory has low bit-error rates by various noises. In this paper, we study the noise of multi-level cell memory, and we propose the channel model of multi-level cell memory.

• Korean Journal of Computational Design and Engineering
• /
• v.9 no.4
• /
• pp.294-305
• /
• 2004

Highly detailed geometric models are rapidly becoming commonplace in computer graphics and other applications. These complex models, which is often represented as complex1 triangle meshes, mainly suffer from the vast memory requirement for real-time manipulation of arbitrary geometric shapes without loss of data. Various techniques have been devised to challenge these problems in views of geometric processing, not a representation scheme. This paper proposes the new mesh structure for the compact representation and the efficient handling of the highly complex models. To verify the compactness and the efficiency, the memory requirement of our representation is first investigated and compared with other existing representations. And then we analyze the time complexity of our data structure by the most critical operation, that is, the enumeration of the so-called one-ring neighborhood of a vertex. Finally, we evaluate some elementary modeling functions such as mesh smoothing, simplification, and subdivision, which is to demonstrate the effectiveness and robustness of our mesh structure in the context of the geometric modeling and processing.

• Economic and Environmental Geology
• /
• v.33 no.1
• /
• pp.41-50
• /
• 2000

Geo-radar survey which has the advantage of high-resolution and relatively fast survey has been widely used for engineering and environmental problems. Three-dimensional effects have to be considered in the interpretation of geo-radar for high-resolution. However, there exists a trouble on the analysis of the three dimensional effects. To solve this problem an efficient three dimension numerical modeling algorithm is needed. Numerical radar modeling in three dimensional case requires large memory and long calculating time. In this paper, a finite difference method time domain solution to Maxwell's equations for simulating electromagnetic wave propagation in three dimensional media was developed to make economic algorithm which requires smaller memory and shorter calculating time. And in using boundary condition Liao absorption boundary. The numerical result of cross-hole radar survey for tunnel is compared with real data. The two results are well matched. To prove application to three dimensional analysis, the results with variation of tunnel's incident angle to survey cross-section and the result when the tunnel is parallel to the cross-section were examined. This algorithm is useful in various geo-radar survey and can give basic data to develop dat processing and inversion program.

• Journal of the Korea Society for Simulation
• /
• v.5 no.2
• /
• pp.25-40
• /
• 1996

Modeling of systems for intensive knowledge-based processing requires a modeling methodology that makes efficient access to the information in huge data base models. The proposed level access mothodology is a modeling approach applicable to systems where data is stored in a hierarchical and modular modules of active memory cells(processor/memory pairs). It significantly reduces the effort required to create discrete event simulation models constructed in hierarchical, modular fashion for above application. Level access mothodology achieves parallel access to models within the modular, hierarchical modules(clusters) by broadcasting the desired operations(e.g. querying information, storing data and so on) to all the cells below a certain desired hierarchical level. Level access methodology exploits the capabilities of object-oriented programming to provide a flexible communication paradigm that combines port-to-port coupling with name-directed massaging. Several examples are given to illustrate the utility of the methodology.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.13 no.4
• /
• pp.277-283
• /
• 2013

In this study, we propose a novel method for modeling dynamic hysteresis. Hysteresis is a widespread phenomenon that is observed in many physical systems. Many different models have been developed for representing a hysteretic system. Among them, the Duhem model is a classical nonlinear dynamic hysteresis model satisfying the properties of hysteresis. The purpose of this work is to develop a novel method that expresses the local dynamics of the Duhem model by a linear system model. Our approach utilizes a certain type of fuzzy system that is based on Takagi-Sugeno (T-S) fuzzy models. The proposed T-S fuzzy Duhem model is achieved by fuzzy blending of the linear system model. A simulated example applied to shape memory alloy actuators, which have typical hysteretic properties, illustrates the applicability of our proposed scheme.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.38 no.3
• /
• pp.237-246
• /
• 1989

A sliding memory covariance circular lattice (SMC-CL) filter and an efficient ARMA modeling method using the SMC-CL filter are presented. At first, SMC-CL filter is derived based on the geometric approach. Then ARMA process is converted into 2 channel AR process, and SMC-CL filter is applied to it. The structure of SMC-CL filter becomes simpler in case of ARMA modeling due to the whiteness of a driving input process. The parameters of ARMR process can be obtained by the Levinson recursions from the PARCOR coefficients of the second channel of the filter. Computer simulations are performed to show the effctiveness of the proposed algorithm.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.14 no.5
• /
• pp.609-614
• /
• 2014

The fringe field effects on the transient characteristics of nano-electromechanical (NEM) memory cells have been discussed by using an analytical model. The influence of fringe field becomes stronger as the size of a cell decreases. By using the proposed model, the dependency of NEM memory transient characteristics on cell parameters has been evaluated.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.10 no.3
• /
• pp.99-108
• /
• 2006

Aerospace applications use pyrotechnic devices with many different functions. Functional shock, safety, overall system cost issue, and availability of new technologies, however, question the continued use of these mechanisms on aerospace applications. Release device is an important example of a task usually executed by pyrotechnic mechanisms. Many aerospace applications like satellite solar panels deployment or weather balloon separation need a release device. Several incidents, where pyrotechnic mechanisms could be responsible for spacecraft failure, have been encouraging new designs for these devices. The Frangibolt is a non explosive device which comprises a commercially available bolt and a small collar made of shape memory alloy (SMA) that replace conventional explosive bolt systems. This paper presents the modeling and simulation of Frangiblot by the change of bolt size and notch geometry. This analysis may contribute to improve the Frangibolt design.