• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.1A
• /
• pp.76-83
• /
• 2009

In this paper, we provide opportunistic scheduling schemes for elastic services in OFDMA systems with fairness constraints for each user. We adopt the network utility maximization framework in which a utility function is defined for each user to represent its level of satisfaction to the service. Since we consider elastic services whose degree of satisfaction depends on its average data rate, we define the utility function of each user as a function of its average data rate. In addition, for fair resource allocation among users, we define fairness requirements of each user by using utility functions. We first formulate an optimization problem for each fairness requirement that aim at maximizing network utility, which is defined as the sum of utilities of users. We then develop an opportunistic scheduling scheme for each fairness requirement by solving the problem using a dual approach and a stochastic sub-gradient algorithm.

• Journal of the Korean Wood Science and Technology
• /
• v.31 no.1
• /
• pp.10-15
• /
• 2003

This study was carried out to investigate a radial distribution of moisture contents(MCs) within a log cross section(LC) during oven-drying of circumferential slices cut from some of domestic hardwoods LCs. For zelkova, the slices from sapwood and from transition part already reached to below hygroscopic range at 30% in average MC, whereas the innermost slice represented about 40% MC. A moisture gradient between the innermost slice and the adjacent slice from a cherry LC became severer with drying times. For castor aralia, the green MC for heartwood slices was approximately 23% higher than that for sapwood slices, and the similar moisture gradient still lasted until average 15% MC during oven-drying. For Chinese mahogany, the MCs of the mixed slices and the innermost slice were lower than those of the slices cut from the other parts during all over drying stage.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.907-917
• /
• 2023

Recently, a self-driving car have applied deep learning technology to advanced driver assistance system can provide convenience to drivers, but it is shown deep that learning technology is vulnerable to adversarial evasion attacks. In this paper, we performed five adversarial evasion attacks, including MI-FGSM(Momentum Iterative-Fast Gradient Sign Method), targeting the object detection algorithm YOLOv5 (You Only Look Once), and measured the object detection performance in terms of mAP(mean Average Precision). In particular, we present a method applying morphology operations for YOLO to detect objects normally by removing noise and extracting boundary. As a result of analyzing its performance through experiments, when an adversarial attack was performed, YOLO's mAP dropped by at least 7.9%. The YOLO applied our proposed method can detect objects up to 87.3% of mAP performance.

• Geomechanics and Engineering
• /
• v.36 no.3
• /
• pp.259-276
• /
• 2024

The undrained shear strength is widely acknowledged as a fundamental mechanical property of soil and is considered a critical engineering parameter. In recent years, researchers have employed various methodologies to evaluate the shear strength of soil under undrained conditions. These methods encompass both numerical analyses and empirical techniques, such as the cone penetration test (CPT), to gain insights into the properties and behavior of soil. However, several of these methods rely on correlation assumptions, which can lead to inconsistent accuracy and precision. The study involved the development of innovative methods using extreme gradient boosting (XGB) to predict the pile set-up component "A" based on two distinct data sets. The first data set includes average modified cone point bearing capacity (q_t), average wall friction (f_s), and effective vertical stress (σ_vo), while the second data set comprises plasticity index (PI), soil undrained shear cohesion (S_u), and the over consolidation ratio (OCR). These data sets were utilized to develop XGBoost-based methods for predicting the pile set-up component "A". To optimize the internal hyperparameters of the XGBoost model, four optimization algorithms were employed: Particle Swarm Optimization (PSO), Social Spider Optimization (SSO), Arithmetic Optimization Algorithm (AOA), and Sine Cosine Optimization Algorithm (SCOA). The results from the first data set indicate that the XGBoost model optimized using the Arithmetic Optimization Algorithm (XGB - AOA) achieved the highest accuracy, with R2 values of 0.9962 for the training part and 0.9807 for the testing part. The performance of the developed models was further evaluated using the RMSE, MAE, and VAF indices. The results revealed that the XGBoost model optimized using XGBoost - AOA outperformed other models in terms of accuracy, with RMSE, MAE, and VAF values of 0.0078, 0.0015, and 99.6189 for the training part and 0.0141, 0.0112, and 98.0394 for the testing part, respectively. These findings suggest that XGBoost - AOA is the most accurate model for predicting the pile set-up component.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.68 no.2
• /
• pp.47-58
• /
• 2023

The quality and yield of crops produced using field cultivation are expected to decrease due to the recent global climate change caused by extreme weather. The plant reproductive stage associated with crop yields is a highly vulnerable period to global warming caused by high temperatures. This study analyzed the adzuki bean's yield properties, antioxidant contents, and pollen viability of adzuki bean (Vigna angularis L.) under high-temperature stress and growth periods in a temperature gradient greenhouse that forms 0 to 4℃ above the outside temperatures. As the main variety of red beans cultivated in Korean farms, the "Arari" red bean was grown in the rain shield greenhouse and the temperature gradient greenhouse from 2021 to 2022 in Milyang, Korea. Compared to 2022, it showed a 0 - 1.0℃ lower temperature during the whole growth period in 2021. However, its average temperatures were 0 - 3.7℃ higher in the vegetative stage and 0.4 - 2.4℃ higher in the anthesis stage in 2021. The lowest yield (6.8 ± 0.7 g) was at the highest temperature (T4: low, 23.6℃; average, 28.5℃; high, 35.8℃) during the anthesis stage in 2021. The temperatures of the mature stage were 1.7 - 3.9℃, which were higher in 2022 than in 2021, although the low temperatures of 2022 were lower than in 2021. The yields of the mature stage in 2022 increased more than in 2021 because of the high temperature of the mature stage. The growth and yield were good at 40.5℃ in the vegetative stage. However, growth was poor when the average temperature was 27.0℃ or higher, and yields decreased during the flowering period. Total polyphenol and flavonoid contents were increased, and the pollen viability was 40.75% in the whole growth period at high temperature (T4: low, 22.9℃; average, 28.8℃; high, 36.9℃). These results showed that the antioxidant levels increased when the antioxidant component was affected at higher temperatures than at normal. In contrast, the pollen viability-related yield decreased as the temperature increased. Our results are the basic data for field growers and the breeding of thermos-tolerance in adzuki beans to prepare for the changeable future climate.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.1 no.1
• /
• pp.17-25
• /
• 1991

The construction details of VGF apparatus with a DM(direct monitoring) furnace for the growth of low defect crystal and characteristics of GaAs crystal grown by this apparatus are described. The average dislocation densities and EL2 concentration of as-grown undoped GaAs along the different solidified fractions exhibit $4{\times}10^{2}-7{\times}10^{3}cm^{-2}$ and $6{\times}10^{14}-4{\times}10^{15}cm^{-3}$, which are less than those observed for liquid encapsulated Czochralski(LEC) or high-pressure vertical gradient freeze(VGF) crystals. These remarkable reduction of the dislocation densities and EL2 concentrations were explained by the lower temperature gradient ($dT/dx-10^{\circ}/cm$) and slower rates of post - growth cooling ($20^{\circ}C/hr:1240-1000^{\circ}C,\;30^{\circ}C/hr:1000-700^{\circ}C$). Also, The Hall mobilities, carrier concentrations show uniform distribution throughtout 80% of the ingot length.

• Smart Media Journal
• /
• v.4 no.2
• /
• pp.26-33
• /
• 2015

In this paper, we propose an efficient automatic pedestrian removal system from a frame in a video sequence. It firstly finds pedestrians from the frame using a Histogram of Oriented Gradient(HOG) / Linear-Support Vector Machine(L-SVM) classifier, searches for proper background patches, and then the patches are used to replace the deleted pedestrians. Background patches are retrieved from the reference video sequence and a modified feather blender algorithm is applied to make boundaries of replaced blocks look naturally. The proposed system, is designed to automatically detect object and generate natural-looking patches, while most existing systems provide search operation in manual. In the experiment, the average PSNR of the replaced blocks is 19.246

• Journal of Ocean Engineering and Technology
• /
• v.13 no.4 s.35
• /
• pp.55-62
• /
• 1999

Cold rolled steel has much plastic strain in the material surface produced by manufacturing process. The strain causes the variation of surface residual stress, in which influences the fatigue behavior under repeated loading. As experimental results, it was confirmed that the behavior of residual stress ${\sigma}_r$, with cycle N consisted of three stages except stress amplitude near fatigue limit in SPCC steel. On the first stage compressive residual stress decreased rapidly, on the second stage gradually, and on the last stage slightly. The relation between ${\sigma}_r$, and log N appeared linear behavior except the early part of cycle ratio $N/N_f$. The average gradient of ${\sigma}_r$, with respect to log N seemed to take a constant value without initial cycle ratio. On the other hand, the $N_f$ line was regressed by the first-order polynomial equation on ${\sigma}_r-log\;N_f$ diagram. Therefore, this study showed that both the gradient of ${\sigma}_r$, with respect to log N and the $N_f$ line was useful in predicting the cycle ratio $N/N_f$.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.26 no.8
• /
• pp.1183-1190
• /
• 2002

An experimental study is performed to investigate the characteristics of near wake behind a circular cylinder with serrated fins using the constant temperature anemometer and through flow visualization. Previous report(Boo et al., 2001) shows that there are three different modes in vortex shedding behavior. This paper is focused on the identification of the physical reasons why the difference iss occured in vortex shedding. The through flow velocity crossing fins decreases as increasing fin height and decreasing fin pitch mainly due to the flow resistence. Vortex shedding is affected strongly by velocity distribution around fin tube, especially by the velocity gradient. The velocity distribution at X/d=0.0 has lower gradient with increasing freestream velocity and fin height and decreasing fin pitch. Those differences in velocity gradients generate different vortex shedding mechanism.

• The Bulletin of The Korean Astronomical Society
• /
• v.37 no.1
• /
• pp.39.1-39.1
• /
• 2012

Nuclear rings in barred galaxies are sites of active star formation (SF). We investigate SF and its feedback effects occurring in barred galaxies, for the first time, using high-resolution grid-based hydrodynamic simulations. The gaseous medium is assumed to be infinitesimally thin, isothermal, and unmagnetized. The SF recipes include a density threshold corresponding to the Jeans condition, a SF efficiency of 1%, and momentum feedback via Type II supernova events together with stellar-wind mass loss. To investigate various environments, we vary the gas sound speed as well as the efficiency of momentum injection in the in-plane direction. We find that when the sound speed is small, the surface density of a ring becomes largely independent of the azimuthal angle, resulting in star-forming regions distributed over the whole length of the ring. When the sound speed is large, on the other hand, the ring achieves the largest density at the contact points between the dust lanes and the ring where SF occurs preferentially, leading to a clear age gradient of star clusters in the azimuthal direction. Since rings shrink with time, a radial age gradient of star clusters naturally develop regardless of sound speed, consistent with observations. SF persists over 200 Myr, with an average rate of ${\sim}1.3M_{\odot}/yr$ similar to observed values. Rings gradually become hostile to SF as they lose gas into stars and turbulent motions dominate.