• Agribusiness and Information Management
• /
• v.1 no.2
• /
• pp.43-60
• /
• 2009

To improve the performance of wide-issue superscalar processors, it is essential to increase the width of instruction fetch and the issue rate. Removal of control hazard has been put forward as a significant new source of instruction-level parallelism for superscalar processors and the conditional branch prediction is an important technique for improving processor performance. Branch mispredictions, however, waste a large number of cycles, inhibit out-of-order execution, and waste electric power on mis-speculated instructions. Hence, the branch predictor with higher accuracy is necessary for good processor performance. In global-history-based predictors like gshare and GAg, many mispredictions come from commit update of the branch history. Some works on this subject have discussed the need for speculative update of the history and recovery mechanisms for branch mispredictions. In this paper, we present a new mechanism for recovering the branch history after a misprediction. The proposed mechanism adds an age_counter to the original predictor and doubles the size of the branch history register. The age_counter counts the number of outstanding branches and uses it to recover the branch history register. Simulation results on the SimpleScalar 3.0/PISA tool set and the SPECINT95 benchmarks show that gshare and GAg with the proposed recovery mechanism improved the average prediction accuracy by 2.14% and 9.21%, respectively and the average IPC by 8.75% and 18.08%, respectively over the original predictor.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.12
• /
• pp.6028-6042
• /
• 2019

Since Sparse Nonnegative Matrix Factorization (SNMF) method can control the sparsity of the decomposed matrix, and then it can be adopted to control the sparsity of facial feature extraction and recognition. In order to improve the accuracy of SNMF method for facial feature recognition, new additive iterative rules based on the improved iterative step sizes are proposed to improve the SNMF method, and then the traditional multiplicative iterative rules of SNMF are transformed to additive iterative rules. Meanwhile, to further increase the sparsity of the basis matrix decomposed by the improved SNMF method, a threshold-sparse constraint is adopted to make the basis matrix to a zero-one matrix, which can further improve the accuracy of facial feature recognition. The improved SNMF method based on the additive iterative rules and threshold-sparse constraint is abbreviated as ASNMF, which is adopted to recognize the ORL and CK+ facial datasets, and achieved recognition rate of 96% and 100%, respectively. Meanwhile, from the results of the contrast experiments, it can be found that the recognition rate achieved by the ASNMF method is obviously higher than the basic NMF, traditional SNMF, convex nonnegative matrix factorization (CNMF) and Deep NMF.

• Journal of the Society of Naval Architects of Korea
• /
• v.46 no.3
• /
• pp.232-248
• /
• 2009

Numerical methodology to solve ship springing problem, which is basically fluid-structure interaction problem, was explored in this study. Solution of this hydroelasticity problem was sought by coupling higher order B-spline Rankine panel method and finite element method in time domain, each of which is introduced for fluid and structure domain respectively. Even though varieties of different combinations in terms of numerical scheme are possible and have been tried by many researchers to solve the problem, no systematic study regarding the characteristics of each scheme has been done so far. Here, extensive case studies have been done on the numerical schemes especially focusing on the iteration method, FE analysis of beam-like structure, handling of forward speed problem and so on. Two different iteration scheme, Newton style one and fixed point iteration, were tried in this study and results were compared between the two. For the solution of the FE-based equation of motion, direct integration and modal superposition method were compared with each other from the viewpoint of its efficiency and accuracy. Finally, calculation of second derivative of basis potential, which is difficult to obtain with accuracy within grid-based method like BEM was discussed.

• Journal of KSNVE
• /
• v.4 no.2
• /
• pp.231-238
• /
• 1994

This paper presents an experimental method to find the vibrational transmission characteristics of structures by using the structural intensity method which is used as the important techniques of active vibration control method. Experimental results are obtained from measurements performed on a structure beam by 2, 3 and 4 position linear accelerometr array (2, 3 and 4 structural intensity : 2, 3 and 4 S.I.) methods at near and farfield conditions. These results are compared with the measurement values of conventional power flow measurement method called input power measurement in order to verify the accuracy of structural intensity methods. To minimize the errors associated with 2, 3 and 4 S.I. methods, the measurement locations were selected by the result of modal analysis and the averaged data by the inter-change of accelerometer array was utilized. In 3 and 4 S.I. methods measured wavenumber instead of theoretical wavenumber was used. This paper shows that measurements of bending wave power flow by using 2, 3 and 4 S.I. methods can give accurate values under general field conditions in structural beam and the accuracy of 2, 3 S.I. methods is higher than 4 S.I. methods. Finally, 2 position linear accelerometer array method is suggested as the practical structural intensity technique.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.19 no.3
• /
• pp.346-352
• /
• 2010

Recently, the high-tech industries, such as the aerospace industry, the auto industry, and the electronics industry, are growing up considerably. Because of that, high machining accuracy and productivity of precision parts have been required. The tooling system is important part in the machining center. HSK tooling system is more suitable than BT tooling system for that of high speed machining center. It is because static stiffness and machining accuracy of HSK tooling system are higher than those of BT tooling system. In this paper, static and dynamic behavior of the HSK tooling System is analyzed according to bearing characteristics and lightweight parts. In order that, three different models of the HSK tooling system are modelled by using a 3D modeling/design program. More stable one in the models of HSK tooling system can be selected by using the FEA(Finite Element Analysis).

• Smart Structures and Systems
• /
• v.17 no.1
• /
• pp.31-43
• /
• 2016

Applications of ultrasonic tomography to concrete structures have been reported for many years. However, practical and effective application of this tool for nondestructive assessment of internal concrete condition is hampered by time consuming transducer coupling that limits the amount of ultrasonic data that can be collected. This research aims to deploy recent developments in air-coupled ultrasonic measurements of solids, described in Part 1 of this paper set, to concrete in order to image internal inclusions. Ultrasonic signals are collected from concrete samples using a fully air-coupled (contactless) test configuration. These air coupled data are compared to those collected using partial semi-contact and full-contact test configurations. Two samples are considered: a 150 mm diameter cylinder with an internal circular void and a prism with $300mm{\times}300mm$ square cross-section that contains internal damaged regions and embedded reinforcement. The heterogeneous nature of concrete material structure complicates the application and interpretation of ultrasonic measurements and imaging. Volumetric inclusions within the concrete specimens are identified in the constructed velocity tomograms, but wave scattering at internal interfaces of the concrete disrupts the images. This disruption reduces defect detection accuracy as compared with tomograms built up of data collected from homogeneous solid samples (PVC) that are described in Part 1 of this paper set. Semi-contact measurements provide some improvement in accuracy through higher signal-to-noise ratio while still allowing for reasonably rapid data collection.

• Transactions of Materials Processing
• /
• v.28 no.4
• /
• pp.212-218
• /
• 2019

In this study, the first drawing die for the production of coil wedge is redesigned in order to enhance properties such as dimensional accuracy, dimensional uniformity, non-magnetism, and residual stress. The equivalent strain distribution is observed to be asymmetric at certain corners of the product and un-filling of material is also observed at the same location, based on the results of FEM simulation for the current drawing process. Additionally, a relatively huge amount of deformation is concentrated on the surface of the reference product leading to an increase in magnetic component and surface residual stress. After re-designing the cross-section of the first drawing step process conformed to relatively higher amount of reduction ratio, reduction of both surface residual stress and the volume fraction of magnetic component could be achieved for the finally-drawn coil wedge product.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.7
• /
• pp.3654-3670
• /
• 2019

This paper proposes an Image Texture Median Filter (ITMF) to analyze and detect Android malware on Drebin datasets. We design a model of "ITMF" combined with Image Processing of Median Filter (MF) to reflect the similarity of the malware binary file block. At the same time, using the MAEVS (Malware Activity Embedding in Vector Space) to reflect the potential dynamic activity of malware. In order to ensure the improvement of the classification accuracy, the above-mentioned features(ITMF feature and MAEVS feature)are studied to train Restricted Boltzmann Machine (RBM) and Back Propagation (BP). The experimental results show that the model has an average accuracy rate of 95.43% with few false alarms. to Android malicious code, which is significantly higher than 95.2% of without ITMF, 93.8% of shallow machine learning model SVM, 94.8% of KNN, 94.6% of ANN.

• Journal of Information Processing Systems
• /
• v.15 no.3
• /
• pp.604-615
• /
• 2019

Single-user spectrum sensing is susceptible to multipath effects, shadow effects, hidden terminals and other unfavorable factors, leading to misjudgment of perceived results. In order to increase the detection accuracy and reduce spectrum sensing cost, we propose an adaptive cooperative sensing strategy based on an estimated signal-to-noise ratio (SNR). Which can adaptive select different sensing strategy during the local sensing phase. When the estimated SNR is higher than the selection threshold, adaptive double threshold energy detector (ED) is implemented, otherwise cyclostationary feature detector is performed. Due to the fact that only a better sensing strategy is implemented in a period, the detection accuracy is improved under the condition of low SNR with low complexity. The local sensing node transmits the perceived results through the control channel to the fusion center (FC), and uses voting rule to make the hard decision. Thus the transmission bandwidth is effectively saved. Simulation results show that the proposed scheme can effectively improve the system detection probability, shorten the average sensing time, and has better robustness without largely increasing the costs of sensing system.

• Analytical Science and Technology
• /
• v.32 no.3
• /
• pp.88-95
• /
• 2019

This study was conducted to establish the analytical method for the determination of cyanide in blood, urine, lung and skin tissues in rats. In order to detect or quantify the sodium cyanide in above biological matrixes, it was derivatized to Pentafluorobenzyl cyanide (PFB-CN) using pentafluorobenzyl bromide (PFB-Br) and then reaction substance was analyzed using gas chromatography mass spectrometer (GC/MS)-SIM (selected ion monitoring) mode. The analytical method for cyanide determination was validated with respect to parameters such as selectivity, system suitability, linearity, accuracy and precision. No interference peak was observed for the determination of cyanide in blank samples, zero samples and lower limit of quantification (LLOQ) samples. The lowest limit detection (LOD) for cyanide was $10{\mu}M$. The linear dynamic range was from 10 to $200{\mu}M$ for cyanide with correlation coefficients higher than 0.99. For quality control samples at four different concentrations including LLOQ that were analyzed in quintuplicate, on six separate occasions, the accuracy and precision range from -14.1 % to 14.5% and 2.7 % to 18.3 %, respectively. The GC/MS-based method of analysis established in this study could be applied to the toxicokinetic study of cyanide on biological matrix substrates such as blood, urine, lung and skin tissues.