• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.1
• /
• pp.10-17
• /
• 2017

We propose classification algorithms for human and dog movement. The proposed algorithms use micro-Doppler signals obtained from humans and dogs moving in four different directions. A two-stage classifier based on a support vector machine (SVM) is proposed, which uses a radial-based function (RBF) kernel and $16^{th}$-order linear predictive code (LPC) coefficients as feature vectors. With the proposed algorithms, we obtain the best classification results when a first-level SVM classifies the type of movement, and then, a second-level SVM classifies the moving object. We obtain the correct classification probability 95.54% of the time, on average. Next, to deal with the difficult classification problem of human and dog running, we propose a two-layer convolutional neural network (CNN). The proposed CNN is composed of six ($6{\times}6$) convolution filters at the first and second layers, with ($5{\times}5$) max pooling for the first layer and ($2{\times}2$) max pooling for the second layer. The proposed CNN-based classifier adopts an auto regressive spectrogram as the feature image obtained from the $16^{th}$-order LPC vectors for a specific time duration. The proposed CNN exhibits 100% classification accuracy and outperforms the SVM-based classifier. These results show that the proposed classifiers can be used for human and dog classification systems and also for classification problems using data obtained from an ultra-wideband (UWB) sensor.

• Proceedings of the Korea Information Assurance Society Conference
• /
• 2004.05a
• /
• pp.103-108
• /
• 2004

A Home Location Register(HLR) database system manages each subscriber's location information, which continuously changes in a cellular network. For this purpose, the HLR database system provides table management, index management, and backup management facilities. In this thesis, I propose using a two-level index method for the mobile directory number(MDN) as a suitable method and a chained bucket hashing method for the electronic serial number(ESN). Both the MDN and the ESN are used as keys in the HLR database system. I also propose an efficient backup method that takes into account the characteristics of HLR database transactions. The retrieval speed and the memory usage of the two-level index method are better than those of the T-tree index method. The insertion and deletion overhead of the chained bucket hashing method is less than that of the modified linear hashing method. In the proposed backup method, I use two kinds of dirty flags in order to solvethe performance degradation problem caused by frequent registration-location operations. I also propose using additional attributes in the HLR database scheme for location information secrecy as a suitable security method.

• International Journal of Concrete Structures and Materials
• /
• v.1 no.1
• /
• pp.83-88
• /
• 2007

This paper discusses a method of measuring transient potential response of a corrosion interface to a small galvanostatic pulse perturbation for a rapid assessment of the corrosion rate of reinforcing steel in concrete structures. Measurements were taken on 100 mm sections of steel bars which were subjected to a wide range of corrosion conditions, from passive steel to actively corroding steel. The duration of the applied galvanostatic pulse was varied between 5s and 180s, and the lateral distance of the point of measurement on the steel bar varied from zero to 400 mm. The result of the electrochemical transient response was investigated using a typical sampling rate of 1 kHz. Analysis of the transient potential response to the applied galvanostatic pulse has enabled the separation of equivalent electronic components so that the components of a series of capacitances and resistances, whose values are dependent on the corrosion condition of the reinforcing steel, could be isolated. The corrosion rate was calculated from a summation of the separate resistive components, which were associated with the corrosion interface, and was compared with the corrosion rate obtained from linear polarization resistance (LPR) method. The results show that the galvanostatic pulse transient technique enables the components of the polarization resistance to be evaluated separately so as to give more reliable corrosion rate values than those obtained from the LPR method. Additionally, this paper shows how the galvanostatic pulse transient response technique can be implemented. An appropriate measurement time for passive and actively corroding reinforcing steel is suggested for the galvanostatic pulse transient response measurements in the field site.

• Journal of the Korea Convergence Society
• /
• v.9 no.6
• /
• pp.53-60
• /
• 2018

In this paper, we described the unknown parameter identification using Dual Extended Kalman Filter for precise control of 2-link manipulator. 2-link manipulator has highly non-linear characteristic with changed parameter thought tasks. The parameter kinds of mass and inertia of system is important to handle with the manipulator robustly. To solve the control problem by estimating the state and unknown parameters of the system through the proposed method. In order to verify the performance of proposed method, we simulate the implementation using Matlab and compare with results of RLS algorithm. At the results, proposed method has a better performance than those of RLS and verify the estimation performance in the parameter estimation.

• Journal of IKEEE
• /
• v.17 no.4
• /
• pp.459-468
• /
• 2013

Under nonlinear characteristics of frames, we propose the frame interpolation using GRNN to enhance the visual picture quality. By full search with block size of 128x128~1x1 to reduce blocky artifact and image overlay, we select the frame having block of minimum error and re-estimate the nonlinear moving vector using GRNN. We compare our scheme with forward(backward) motion compensation, bidirectional motion compensation when the object movement is large or the object image includes zoom-in and zoom-out or camera focus has changed. Experimental results show that the proposed method provides better performance in subjective image quality compared to conventional MCFI methods.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.3
• /
• pp.199-203
• /
• 2016

A brain-computer interface or BCI provides an alternative method for acting on the world. Brain signals can be recorded from the electrical activity along the scalp using an electrode cap. By analyzing the EEG, it is possible to determine whether a person is thinking about his/her hand or foot movement and this information can be transferred to a machine and then translated into commands. However, we do not know which information relates to motor imagery and which channel is good for extracting features. A general approach is to use all electronic channels to analyze the EEG signals, but this causes many problems, such as overfitting and problems removing noisy and artificial signals. To overcome these problems, in this paper we used a new optimization method called the Fruit Fly optimization algorithm (FOA) to select the best channels and then combine them with CSP method to extract features to improve the classification accuracy by linear discriminant analysis. We also used particle swarm optimization (PSO) and a genetic algorithm (GA) to select the optimal EEG channel and compared the performance with that of the FOA algorithm. The results show that for some subjects, the FOA algorithm is a better method for selecting the optimal EEG channel in a short time.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.11
• /
• pp.27-32
• /
• 1999

Metal-ferroelectric-semiconductor devices by susing rapid thermal annealed $LiNbO_3/Si$(100) structures were fabricated and demonstrated nonvolatile memory operations. The estimated field-effect electron mobility and transconductance on a linear region of the fabricated FET were about $600cm^2/V{\cdot}s$ and 0.16mS/mm, respectively. The ID-VG characteristics of MFSFET's showed a hysteresis loop due to the ferroelectric nature of the $LiNbO_3 films. The drain current of the on state was more than 4 orders of magnitude larger than the off state current at the same read gate voltage of 0.5V, which means the memory operation of the MFSFET. A write voltage as low as ${\pm}3V$, which is applicable to low power integrated circuits, was used for polarization reversal. The ferroelectric capacitors showed no polarization degradation up to $10^{10}$ switching cycles with the application of symmetric bipolar voltage pulse (peak-to-peak 6V, 50% duty cycle) of 500kHz.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.3
• /
• pp.202-209
• /
• 2014

In this paper, we analyzed the effects by multi-stacking superstrates over the circular-polarized patch antenna. The previous works considered a single-layered superstrate or a superstrate with multiple layers, and did not almost consider the axial ratio at the performance analysis. First, the effect of center frequency shift is analyzed by the variation of air-gap height between patch antenna and superstrate. The center frequency is down-shifted at the smaller air-gap height and has almost the same frequency as patch antenna at the air-gap height of $005{\lambda}_0$. Second, the antenna performance is analyzed by multi-stacking superstrates with the air-gap height of $005{\lambda}_0$. As the number of multi-stacked superstrates increase, antenna gain has a linear increase and axial ratio is exponentially deteriorated. In addition, it has also been observed that the antenna performance has the same trend with the number of multi-stacked superstrates as the thickness of superstrate increases. Finally, we confirmed that it is possible to design the CP patch antenna with the scalable gain and less than 3dB axial ratio by stacking the superstrate.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.16 no.12 s.103
• /
• pp.1194-1205
• /
• 2005

This paper proposes a uni-planar elliptical element UWB antenna with band-notched characteristic and modified pound plane fed by CPW. The antenna achieves VSWR below 2 for UWB band(3.1 GHz${\~}$10.6 GHz). In addition, a band-notched characteristic is achieved by inserting a horizontal slot on the radiation element to avoid the interference with 5 GHz(5.15 GHz${\~}$5.825 GHz) band limited by IEEE 802.1la. The antenna has a thin profile, compact size, and ease of manufacture by adopting a CPW feed structure without any additional background plane. Measured data show that the proposed antenna has good return loss below -10 dB, about 2.1 dB${\~}$4.75 dB maximum gain over the bandwidth, omni-directional radiation patterns, linear phase response.

• Journal of Korea Foundry Society
• /
• v.34 no.3
• /
• pp.100-106
• /
• 2014

In recent years, Mg and its alloys have attracted a great deal of attention due to their low density, relatively excellent castability, and straightforward recyclability. Mg alloys have been widely applied to various industrial fields, and are representatively used in automotive and electronic parts. According to previous researches, the electrical conductivity of Mg alloys greatly decreases with increasing Al content. However, with the addition of Zn and/or Cu, the electrical conductivity of Mg alloys is maintained or slightly increased, and improved mechanical properties are obtained as well. On this basis, Mg-Zn-Cu alloys have been investigated in the present study with a focus on the effect of adding Zn and Cu on the electrical conductivity. The Zn and Cu contents ranged from 4 to 6wt.% and 0 to 1.5wt.%, respectively. Ternary Mg-Zn-Cu alloys have been prepared by gravity casting in a steel mold. In the as-casting condition, the electrical conductivity of Mg-Zn-Cu alloys showed a linear increasing trend with decreasing Zn and increasing Cu contents. Furthermore, impact values of Zn = -1.5 and Cu = 2.5 were determined for these alloys by electrical conductivity tests.