• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.4
• /
• pp.161-166
• /
• 2018

In this paper, we propose a method for the random selection of pooling operations for the regularization and reduction of cross validation in convolutional neural networks. The pooling operation in convolutional neural networks is used to reduce the size of the feature map and for its shift invariant properties. In the existing pooling method, one pooling operation is applied in each pooling layer. Because this method fixes the convolution network, the network suffers from overfitting, which means that it excessively fits the models to the training samples. In addition, to find the best combination of pooling operations to maximize the performance, cross validation must be performed. To solve these problems, we introduce the probability concept into the pooling layers. The proposed method does not select one pooling operation in each pooling layer. Instead, we randomly select one pooling operation among multiple pooling operations in each pooling region during training, and for testing purposes, we use probabilistic weighting to produce the expected output. The proposed method can be seen as a technique in which many networks are approximately averaged using a different pooling operation in each pooling region. Therefore, this method avoids the overfitting problem, as well as reducing the amount of cross validation. The experimental results show that the proposed method can achieve better generalization performance and reduce the need for cross validation.

• Korean Journal of Optics and Photonics
• /
• v.12 no.3
• /
• pp.240-249
• /
• 2001

A high speed ultra-compact lens with a diffractive optical element (DOE) is designed, which can be applied to mobile communication devices such as IMT2000, PDA, notebook computer, etc. The designed hybrid lens has sufficiently high performance of less than f/2.2, compact size of 3.3 mm (1st surf. to image), and wide field angle of more than 30 deg. compared with the specifications of a single lens. By proper choice of the aspheric and DOE surface which has very large negative dispersion, we can correct chromatic and high order aberrations through the optimization technique. From Seidel third order aberration theory and Sweatt modeling, the initial data and surface configurations, that is, the combination condition of the DOE and the aspherical surface are obtained. However, due to the consideration of diffraction efficiency of a DOE, we can choose only four cases as the optimization input, and present the best solution after evaluating and comparing those four cases. On the other hand, we also report dramatic improvement in optical performance by inserting another refractive lens (so-called, field flattener), that keeps the refractive power of an original DOE lens and makes the petzval sum zero in the original DOE lens system. ystem.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.19 no.1
• /
• pp.23-32
• /
• 2008

In the VHF range, active antennas are widely used for wideband applications due to their small size. Active antenna consists of antenna elements and amplifiers, which are directly connected to each other. Gain and noise-figure characteristics are very important for good sensitivity performance, because it is located at the front end of a receiving system. In this study, we developed an active dipole antenna with 5:1 bandwidth(100${\sim}$500 MHz), which consists of a dipole antenna and a P-HEMT amplifier. To obtain required performances, the antenna and the amplifier should be designed simultaneously. In order for that, we introduced an equivalent port concept to model the 1-port dipole antenna as an equivalent 2-port system. Using the proposed equivalent port, the performance of the active dipole antenna was simulated by the ADS. In order to measure the gain and noise-figure characteristics of the antenna, we utilized the same concept of the two-port equivalent impedance model. The measurement results for typical gain, NF and VSWR in the required frequency band were 8dBi, 9dB and 1.7:1, respectively. The radiation patterns at the principal planes were same as the typical radiation pattern of a dipole antenna. By comparing the simulation results with measured ones, it is confirmed that the proposed methods works well.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.27 no.6
• /
• pp.566-576
• /
• 2016

Inverse synthetic aperture radar(ISAR) image generated from bistatic radar(Bi-ISAR) represents two-dimensional scattering distribution of a target, and the Bi-ISAR can be used for bistatic target identification. However, Bi-ISAR has large variability in scattering mechanisms depending on bistatic configurations and do not represent exact range-Doppler information of a target due to inherent distortion. Thus, an efficient training DB construction is the most important factor in target identification using Bi-ISARs. Recently, a database construction method based on realistic flight scenarios of a target, which provides a reliable identification performance for the monostatic target identification, was applied to target identification using high resolution range profiles(HRRPs) generated from bistatic radar(Bi-HRRPs), to construct efficient training DB under bistatic configurations. Consequently, high identification performance was achieved using only small amount of training Bi-HRRPs, when the target is a considerable distance away from the bistatic radar. Thus, flight scenarios based training DB construction is applied to target identification using Bi-ISARs. Then, the capability and efficiency of the method is analyzed.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.20 no.11
• /
• pp.1161-1170
• /
• 2009

In this paper, dielectric, drude model and ideal metamaterial are applied to core of A-Sandwich radome and each radome is analyzed using recursive method in Ku band. The main parameters of radome performance are insertion loss, insertion phase delay and depolarization. In case of ideal metamaterial, the radome using ideal metamaterial dose not generate depolarization because insertion loss, insertion phase delay and loss for incidence angle of wave do not happened. If circular polarization wave is incident on radome with meta material, transmitted wave also keeps circular polarization. In case of the dispersive metamaterial, the performance of radome using dispersive metamaterial is better than it of radome using dielectric in a part of frequency band. From these results, it is showed that metamaterial can be applied to various radome structure.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.5
• /
• pp.495-503
• /
• 2014

In recent days, a study of the intrusion detection system collecting and analyzing network data, packet or logs, has been actively performed to response the network threats in computer security fields. In particular, Bayesian network has advantage of the inference functionality which can infer with only some of provided data, so studies of the intrusion system based on Bayesian network have been conducted in the prior. However, there were some limitations to calculate high detection performance because it didn't consider the problems as like complexity of the relation among network packets or continuos input data processing. Therefore, in this paper we proposed two methodologies based on K-menas clustering to improve detection rate by reforming the problems of prior models. At first, it can be improved by sophisticatedly setting interval range of nodes based on K-means clustering. And for the second, it can be improved by calculating robust CPT through applying weighted-leaning based on K-means clustering, too. We conducted the experiments to prove performance of our proposed methodologies by comparing K_WTAN_EM applied to proposed two methodologies with prior models. As the results of experiment, the detection rate of proposed model is higher about 7.78% than existing NBN(Naive Bayesian Network) IDS model, and is higher about 5.24% than TAN(Tree Augmented Bayesian Network) IDS mode and then we could prove excellence our proposing ideas.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.49 no.7
• /
• pp.39-47
• /
• 2012

The combination of the broadband property and the wide area coverage of satellite communications enables high speed transmission. Every user in the region under the satellite beam coverage can tranceiver and one can simultaneously communicate with multiple users. For these reasons, it is one of commendable telecommunication networks for information transfer. Since the satellite communications use open channels, it is likely to cause jamming with unwanted interference signals. In the thesis, APSK (Amplitude Phase Shift Keying) is employed, which is recommended for DVB-S2 due to high-speed transmission and excellent bandwidth efficiency. For obtaining reliable communication under the jamming environments, the communication satellite transponder rests on the polyphase filter bank structure, which enables switching among the subchannels and gain control on each subchannel, resulting in effectively eliminating jamming. Furthermore, the nulling scheme, one of the various anti-jamming approaches, is investigated, in which unwanted jamming signals are eliminated in the frequency domain after passing through the analysis part of the polyphase filter bank. The performance of the nulling scheme is evaluated for tone jamming and partial band jamming in terms of BER and EVM. The simulation results indicate that the nulling scheme improve the BER and EVM performance over the case without any anti-jamming approach.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.27 no.2
• /
• pp.198-207
• /
• 2016

Bistatric radar can perform detection and identification for stealth targets that are rarely detected by the conventional monostatic radar. However, high resolution range profile(HRRP) generated from the received signal in the bistatic radar cannot show exact range information of the target because the bistatic geometry lead to the distortions of the bistatic HRRP. In addition, electromagnetic scattering mechanisms of the target are varied depending on the bistatic geometry. Thus, efficient database construction is a crucial factor to achieve successful classification capability in bistatic target identification. In this paper, a database construction method based on realistic flight scenarios of a target, which provides a reliable identification performance for the monostatic radar, is applied to bistatic target identification. Then, the capability and efficiency of the method is analyzed. Simulation results show that reliable identification performance can be achieved using the database construction based on the flight scenarios when the target is a considerable distance away from the bistatic radar.

• Journal of the Korean Ceramic Society
• /
• v.43 no.12 s.295
• /
• pp.812-822
• /
• 2006

This study amis to investigate the functional analysis of anode and cathode materials in Anode supported Solid Oxide Fuel Cell. The concentration polarization of single cell was investigated with CFD (Computational Fluid Dynamics) method for the case of the different morphology by using four types of unit cell and discussed to reduce the concentration polarization. The concentration polarization at anode side effected the voltage loss in Anode supported Solid Oxide Fuel Cell and increased contact areas between fuel gas and anode side could reduce the concentration polarization. For intermediate temperature operation, Anode-supported single cells with thin electrolyte layer of YSZ (Yttria-Stabilized Zirconia) were fabricated and short stacks were built and evaluated. We also developed diesel and methane autothermal reforming (ATR) reactors in order to provide fuels to SOFC stacks. Influences of the $H_2O/C$ (steam to carbon ratio), $O_2/C$ (oxygen to carbon ratio) and GHSV (Gas Hourly Space Velocity) on performances of stacks have been investigated. Performance of the stack operated with a diesel reformer was lower than with using hydrogen as a fuel due to lower Nernst voltage and carbon formation at anode side. The stack operated with a natural gas reformer showed similar performances as with using hydrogen. Effects of various reformer parameters such as $H_2O/C$ and $O_2/C$ were carefully investigated. It is found that $O_2/C$ is a sensitive parameter to control stack performance.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.36 no.1
• /
• pp.31-36
• /
• 2012

Military vehicles, compared to conventional vehicles, require higher driving performance, quieter operation, and longer driving distances with minimal fuel supplies. The series hybrid electric vehicle can be driven with no noise and has high initial startup performance, because it uses only a traction motor that has a high startup torque to drive the vehicle. Moreover, the fuel economy can be improved if the vehicle is hybridized. In series hybrid electric vehicles, the electric generation system, which consists of an engine and a generator, supplies electric energy to a battery or traction motor depending on the vehicle driving state and battery state of charge (SOC). The control strategy determines the operation of the generation system. Thus, the fuel economy of the series hybrid electric vehicle relies on the control strategy. In this study, thermostat, power-follower, and combined strategies were compared, and a 37% improvement in the fuel economy was implemented using the combined control strategy suggested in this study.