• IEMEK Journal of Embedded Systems and Applications
• /
• v.2 no.3
• /
• pp.145-154
• /
• 2007

Dynamic update of kernel can change kernel functionality and fix bugs in runtime. Dynamic update is important because it leverages availability, reliability and flexibility of kernel. An instruction-granularity update technique has been used for dynamic update. However, it is difficult to apply update technique for a commodity operating system kernel because development and maintenance of update code must be performed with assembly language. To overcome this difficulty, we design the function-granularity dynamic update system which uses high-level language such as C language. The proposed update system makes the development and execution of update convenient by providing the development environment for update code which is same for kernel development. We implement this system for Linux and demonstrate an example of update for do_coredump() function which is reported it has a vulnerable point for security. The update was successfully executed.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.32 no.3
• /
• pp.225-231
• /
• 2014

When a spatial regression model that uses kernel density values as a dependent variable is applied to retail business data, a unique model cannot be selected because kernel density values change following kernel bandwidths. To overcome this problem, this paper suggests how to use the point pattern analysis, especially the L-index to select a unique spatial regression model. In this study, kernel density values of retail business are computed by the bandwidth, the distance of the maximum L-index and used as the dependent variable of spatial regression model. To test this procedure, we apply it to meeting room business data in Seoul, Korea. As a result, a spatial error model (SEM) is selected between two popular spatial regression models, a spatial lag model and a spatial error model. Also, a unique SEM based on the real distribution of retail business is selected. We confirm that there is a trade-off between the goodness of fit of the SEM and the real distribution of meeting room business over the bandwidth of maximum L-index.

• ETRI Journal
• /
• v.15 no.2
• /
• pp.35-51
• /
• 1993

This paper presents function approximation based on nonparametric estimation. As an estimation model of function approximation, a three layered network composed of input, hidden and output layers is considered. The input and output layers have linear activation units while the hidden layer has nonlinear activation units or kernel functions which have the characteristics of bounds and locality. Using this type of network, a many-to-one function is synthesized over the domain of the input space by a number of kernel functions. In this network, we have to estimate the necessary number of kernel functions as well as the parameters associated with kernel functions. For this purpose, a new method of parameter estimation in which linear learning rule is applied between hidden and output layers while nonlinear (piecewise-linear) learning rule is applied between input and hidden layers, is considered. The linear learning rule updates the output weights between hidden and output layers based on the Linear Minimization of Mean Square Error (LMMSE) sense in the space of kernel functions while the nonlinear learning rule updates the parameters of kernel functions based on the gradient of the actual output of network with respect to the parameters (especially, the shape) of kernel functions. This approach of parameter adaptation provides near optimal values of the parameters associated with kernel functions in the sense of minimizing mean square error. As a result, the suggested nonparametric estimation provides an efficient way of function approximation from the view point of the number of kernel functions as well as learning speed.

• Journal of Biosystems Engineering
• /
• v.15 no.2
• /
• pp.99-109
• /
• 1990

Rough rice is subjected to a series of static and dynamic forces during mechanical harvesting, handling and processing operations. The mechanical properties such as bioyield point, compressive strength, and deformations at the bioyield point and rupture point are important engineering data needed to develop processing machines and to determine reasonable operating conditions for these machines. The objectives of this study were to determine the mechanical properties of the rough rice kernel at loading rate of 0.664 mm/min and 1.673 mm/min and at various moisture contents, and to examine the effect of the moisture content and the loading rate on these mechanical properties. The follwing results were obtained from the study. 1. Bioyield point, rupture point, bioyield strength and ultimate strength of the rough rice kernel generally decreased in magnitude with an increase in moisture content. A little larger values of these mechanical properties were obtained at the higher loading rate. The rough rice variety and the loading rate affected significantly these mechanical properties at low moisture content, but not at the higher moisture levels. 2. Bioyield point of the sample grains varied from 20 to 80 N, and rupture point varied from 45 to 130N. Bioyield point for Japonica-type rough rice was a little higher than that for Indica-type rough rice, but there were little differnces in rupture point between two types of rough rice. 3. Bioyield strength and ultimate strength of the Japonica-type rough rice varied from 10 MPa. to 39 MPa., and from 13 MPa. to 45 MPa. respectively. Those of the Indica-type rough rice varied from 12 MPa. to 42 MPa., and from 15 MPa. to 53 MPa. respectively. 4. Deformations at bioyield point and rupture point ranged from 0.18 mm/min to 0.26 mm, and from 0.28 mm to 0.53 mm respectively. These deformations decreased with an increase in moisture content up to moisture content of approximately 17% (w.b.) and increased again thereafter. 5. Regression equations were developed to predict these mechanical properties for the rough rice kernel as a function of moisture content.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1996.11a
• /
• pp.774-778
• /
• 1996

Developed in this research is a surface modeling kernel for various CAD/CAM applications. Its internal surface representations are rational parametric polynomials, which are generalizations of nonrational Bezier, Ferguson, Coons and NURBS surface, and are very fast in evaluation. The kernel is designed under the OOP concepts and coded in C++ on PCs. The present implementation of the kernel supports surface construction methods, such as point data interpolation, skinning, sweeping and blending. It also has NURBS conversion routines and offers the IGES and ZES format for geometric information exchange. It includes some geometric processing routines, such as surface/surface intersection, curve/surface intersection, curve projection and so forth. We are continuing to work with the kernel and eventually develop a B-Rep based solid modeler.

• Journal of Information Processing Systems
• /
• v.17 no.6
• /
• pp.1157-1169
• /
• 2021

Modern graphics processor unit (GPU) architectures offer significant hardware resource enhancements for parallel computing. However, without software optimization, GPUs continuously exhibit hardware resource underutilization. In this paper, we indicate the need to alter different warp scheduler schemes during different kernel execution periods to improve resource utilization. Existing warp schedulers cannot be aware of the kernel progress to provide an effective scheduling policy. In addition, we identified the potential for improving resource utilization for multiple-warp-scheduler GPUs by sharing stalling warps with selected warp schedulers. To address the efficiency issue of the present GPU, we coordinated the kernel-aware warp scheduler and warp sharing mechanism (KAWS). The proposed warp scheduler acknowledges the execution progress of the running kernel to adapt to a more effective scheduling policy when the kernel progress attains a point of resource underutilization. Meanwhile, the warp-sharing mechanism distributes stalling warps to different warp schedulers wherein the execution pipeline unit is ready. Our design achieves performance that is on an average higher than that of the traditional warp scheduler by 7.97% and employs marginal additional hardware overhead.

• Proceedings of the Korean Statistical Society Conference
• /
• 2002.05a
• /
• pp.1-5
• /
• 2002

We consider the problem of optimal bandwidth choice for nonparametric classification, based on kernel density estimators, where the problem of interest is distinguishing between two univariate distributions. When the densities intersect at a single point, optimal bandwidth choice depends on curvatures of the densities at that point. The problem of empirical bandwidth selection and classifying data in the tails of a distribution are also addressed.

• Journal of applied mathematics & informatics
• /
• v.8 no.3
• /
• pp.731-742
• /
• 2001

A Gaussian smoothing algorithm obtained from a cascade of convolutions with a seven-point kernel is described. We prove that the change of local sums after applying our algorithm to sinusoidal signals is reduced to about two thirds of the change by the binomial coefficients. Hence, our seven point kernel is better than the binomial coefficients when trend curves are needed to be generated. We also prove that if our Gaussian convolution is applied to sinusoidal functions, the amplitude of higher frequencies reduces faster than the lower frequencies and hence that it is a low pass filter.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2002.10a
• /
• pp.599-603
• /
• 2002

We propose a new meshfree method to be called the fast moving least square reproducing kernel collocation method(FCM). This methodology is composed of the fast moving least square reproducing kernel(FMLSRK) approximation and the point collocation scheme. Using point collocation makes the meshfree method really come true. In this paper, FCM Is shown to be a good method at least to calculate the numerical solutions governed by second order elliptic partial differential equations with geometric singularity or geometric multi-scales. To treat such problems, we use the concept of variable dilation parameter.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.1
• /
• pp.79-87
• /
• 2012

In the case that the probability density function has a discontinuity point, Huh (2002) estimated the location and jump size of the discontinuity point based on the difference between the right and left kernel density estimators using the one-sided kernel function. In this paper, we consider the cross-validation, made by the right and left maximum likelihood cross-validations, for the bandwidth selection in order to estimate the location and jump size of the discontinuity point. This method is motivated by the one-sided cross-validation of Hart and Yi (1998). The finite sample performance is illustrated by simulated example.