• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.47 no.1
• /
• pp.28-37
• /
• 2010

In this paper, the IC (Integrated Circuit) for multi-channel asynchronous communication was designed by using FPGA and VHDL language. The existing chips for asynchronous communication that has been used commercially are composed of one to two channels. Therefore, when communication system with two channels or more is made, the cost becomes high and it becomes complicated for communication system to be realized and also has very little buffer, load that is placed into Microprocessor increases heavily in case of high speed communication or transmission of high-capacity data. The designed IC was improved the function and performance of communication system and reduced costs by designing 8 asynchronous communication channels with only one IC, and it has the size of transmitter/receiver buffer with 256 bytes respectively and consequently high speed communication became possible. To detect errors between communications, it was designed with digital filter and check-sum logic and channel MUX logic so that the malfunction can be prevented and errors can be detected more easily and input/output port regarding each communication channel can be used flexibly and consequently the reliability of system was improved. It was composed and simulated logic of VHDL described by using Cyclone II Series EP2C35F672C8 and QuartusII V8.1 of ALTERA company. In order to show the performance of designed IC, the test was conducted successfully in QuartusII simulation and experiment and the excellency was compared with TL16C550A of TI (Texas Instrument) company and ATmegal28 general-purpose micro controller of ATMEL company that are used widely as chips for asynchronous communication.

• Journal of IKEEE
• /
• v.22 no.3
• /
• pp.822-828
• /
• 2018

Recently, energy issues such as massive blackout due to increase in power consumption have been emerged, and it is necessary to improve the accuracy of prediction of power consumption as a solution for these problems. In this study, we investigate the difference between the actual power consumption and the predicted power consumption through the deep learning- based power consumption forecasting experiment, and the possibility of adjusting the power reserve ratio. In this paper, the prediction of the power consumption based on the deep learning can be used as a basis to reduce the power reserve ratio so as not to excessively produce extra power. The deep learning method used in this paper uses a learning model of long-short-term-memory (LSTM) structure that processes time series data. In the computer simulation, the generated power consumption data was learned, and the power consumption was predicted based on the learned model. We calculate the error between the actual and predicted power consumption amount, resulting in an error rate of 21.37%. Considering the recent power reserve ratio of 45.9%, it is possible to reduce the reserve ratio by 20% when applying the power consumption prediction algorithm proposed in this study.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.15 no.9
• /
• pp.896-905
• /
• 2004

In this paper, the design, fabrication and experiment on a planar array antenna with a flat-topped radiation pattern for a mobile base station antenna were described. The current distribution of an antenna aperture, which is easily realizable in a feeding network compared with the conventional one of sin(x)/x was optimized for shaping a desired flat-topped radiation pattern. The planar array antenna designed in this paper has a rectangular lattice and is composed of array elements of 16${\times}$8. Each radiating element, which is a microstrip element fed coaxially, has a linear vertical polarization and the feed network which use a Wilkinson power divider and a 180$^{\circ}$ ring hybrid coupler as a base element is designed. The flat-topped radiation pattern with 90$^{\circ}$ is shaped by 16 array elements with the element spacing of 0.55 λ$_{ο}$ in the azimuth plane, and the normal radiation pattern with 10$^{\circ}$ is shaped by 8 array elements with the element spacing of 0.65 λ$_{ο}$ in the elevation plane. Also, the planar array antenna is symmetrically divided into four parts. It consists of one hundred-twenty-eight radiating elements, thirty-two 1-4 column dividers, low 1-8 row dividers and one 1-4 input power divider. In order to verify electrical performances of the planar way antenna proposed in this paper, the experimental breadboard operated in tile band of 1.92～2.17 GHz(IMT2000 band) was fabricated, and its experimental results were a good agreement with simulation ones.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.25 no.11
• /
• pp.1172-1182
• /
• 2014

In these days, many kinds of goods are more light and more integrated. As frequency range of mobile applications have increased to improve performance of antenna furthermore, EMI(ElectroMagnetic Interference) problem has frequently caused by disturbance of antenna in device which aggravates other circuit. This paper proposes a technique for the prediction of noise power to the transmission line from antenna located near the line. Although noise power transferred to transmission line is varied by source impedance of antenna and load impedance of transmission line basically, the power magnitude can be presented in a square form of S-parameter between antenna and transmission line due to small variation of transferred power. For this reason, we can use the index expressed the transferred power varied along geometrical shapes of transmission line. As a result, big difference is occurred along location of antenna especially the bended line. And this such experiment is correspond with simulation, these results have meaning physically considering electromagnetic field distribution in near and far field. HFSS of Ansys and CPW with ground is used in this paper.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.629-635
• /
• 2019

In Deep Learning models derivative is implemented by error back-propagation which enables the model to learn the error and update parameters. It can find the global (or local) optimal points of parameters even in the complex models taking advantage of a huge improvement in computing power. However, deliberately generated data points can 'fool' models and degrade the performance such as prediction accuracy. Not only these adversarial examples reduce the performance but also these examples are not easily detectable with human's eyes. In this work, we propose the method to detect adversarial datasets with random noise addition. We exploit the fact that when random noise is added, prediction accuracy of non-adversarial dataset remains almost unchanged, but that of adversarial dataset changes. We set attack methods (FGSM, Saliency Map) and noise level (0-19 with max pixel value 255) as independent variables and difference of prediction accuracy when noise was added as dependent variable in a simulation experiment. We have succeeded in extracting the threshold that separates non-adversarial and adversarial dataset. We detected the adversarial dataset using this threshold.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.591-596
• /
• 2019

In this paper, we implemented a deep learning operation structure with less influence of local PC performance. In general, the deep learning model has a large amount of computation and is heavily influenced by the performance of the processing PC. In this paper, we implemented deep learning operation using AWS and streaming server to reduce this limitation. First, deep learning operations were performed on AWS so that deep learning operation would work even if the performance of the local PC decreased. However, with AWS, the output is less real-time relative to the input when computed. Second, we use streaming server to increase the real-time of deep learning model. If the streaming server is not used, the real-time performance is poor because the images must be processed one by one or by stacking the images. We used the YOLO v3 model as a deep learning model for performance comparison experiments, and compared the performance of local PCs with instances of AWS and GTX1080, a high-performance GPU. The simulation results show that the test time per image is 0.023444 seconds when using the p3 instance of AWS, which is similar to the test time per image of 0.027099 seconds on a local PC with the high-performance GPU GTX1080.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.35 no.9
• /
• pp.939-946
• /
• 2011

In thermal power generation companies, the recycling of refined ash (LOI < 6%) obtained from a PC-firing furnace is beneficial for the companies, e.g., it can be used for making lightweight aggregates. However, ash having a high LOI, which cannot be reused, is still buried in the ground. To obtain refined ash, the re-burning of high-LOI ash (LOI > 6%) in a PC-firing furnace can be an alternative. In this study, a numerical analysis was performed to demonstrate the effects of ash re-burning. An experimental constant value was decided by TGA (thermo-gravimetric analysis), and a DTF (drop-tube furnace) was used in the experiment for calculating the combustion of ash. On the basis of the trajectory of the moving particles of coal and ash, it was concluded that supplying ash near the burner, which is located high above the ground, is appropriate. On the basis of numerical results, it was concluded that an ash supply rate of 6 ton/h is suitable for combustion, without affecting the PC-firing boiler.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.1
• /
• pp.81-87
• /
• 2010

The coverage ability is one of essential techniques for the Robotic Vacuum Cleaner (RVC). Most of the RVCs rely on random or regular pattern movement to cover a target space due to the technical difficulties to implement localization and map and constraints of hardwares such as controller and sensors. In this paper, we consider two main issues which are low computational load and using sensors with very limited sensing capabilities. First, in our approach, computing procedures to build map and detect the RVC's position are minimized by simplifying data obtained from sensors. To reduce computational load, it needs simply presenting an environment with objects of various shapes. Another isuue mentioned above is regarded as one of the most important problems in our approach, because we consider that many RVCs use low-cost sensor systems such as an infrared sensor or ultrasonic sensor with limited capabilities in limited range, detection uncertainty, measurement noise, etc. Methods presented in this paper are able to apply to general RVCs equipped with these sensors. By both simulation and real experiment, we evaluate our method and verify that the proposed method guarantees a complete coverage.

• Asian-Australasian Journal of Animal Sciences
• /
• v.31 no.1
• /
• pp.40-46
• /
• 2018

Objective: To examine the effects of Rhodobacter sphaeroides (R. sphaeroides) supplementation as a direct-fed microbial (DFM) on rumen fermentation in dairy cows and on coenzyme Q10 (CoQ10) transition into milk, an in vitro rumen simulation batch culture and an in vivo dairy cow experiment were conducted. Methods: The characteristics of in vitro ruminal fermentation were investigated using rumen fluids from six cannulated Holstein dairy cows at 2 h post-afternoon feeding. A control treatment was included in the experiments based on a typified total mixed ration (TMR) for lactating dairy cows, which was identical to the one used in the in vivo study, plus R. sphaeroides at 0.1%, 0.3%, and 0.5% TMR dry matter. The in vivo study employed six ruminally cannulated lactating Holstein cows randomly allotted to either the control TMR (C-TMR) treatment or to a diet supplemented with a 0.5% R. sphaeroides culture (S-TMR, dry matter basis) ad libitum. The presence of R. sphaeroides was verified using denaturing gradient gel electrophoresis (DGGE) applied to the bacterial samples obtained from the in vivo study. The concentration of CoQ10 in milk and in the supernatant from the in vitro study was determined using high performance liquid chromatography. Results: The results of the in vitro batch culture and DGGE showed that the concentration of CoQ10 significantly increased after 2 h of R. sphaeroides supplementation above 0.1%. When supplemented to the diet of lactating cows at the level of 0.5%, R. sphaeroides did not present any adverse effect on dry matter intake and milk yield. However, the concentration of CoQ10 in milk dramatically increased, with treated cows producing 70.9% more CoQ10 than control cows. Conclusion: The CoQ10 concentration in milk increased via the use of a novel DFM, and R. sphaeroides might be used for producing value-added milk and dairy products in the future.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.5
• /
• pp.69-75
• /
• 2001

In this paper, a fast and easy impedance matching method, which could give the impedance matching component for the general 1 or 2-port network was introduced. First, the entire network structure was defined which consists of the network part to be matched and the impedance matching part composed of inductors and capacitors. Next, the transmission matrix and input and output impedances of the entire network from the terminal impedance conditions were calculated, then the exact solutions for the matching components were obtained. To verify the efficiency of this method, this method was applied to the CDMA If band withdrawal weighted SAW transversal filter, and investigated the effects of the impedance matching before and after, through the simulation and experiment. As the result, the performance of a fractional bandwidth of 1.2%, insertion loss of 29 dB, and VSWR of 80 have improved to a factional bandwidth of 1.8％, insertion loss of 9 dB, VSWR of 3 at 85.38 MHz center frequency. The result shows that this impedance matching method could be used in the SAW devices and other types of 1 or 2-port network.