상태 방정식을 이용한 능동 R 필터의 특성 해석 및 합성에 관한 연구

오세호,김부균 崇田大學校 1983 論文集 Vol.13 No.1

Characteristics of active - R filters are expressed by dynamical equations using one-pole roll-off model for the operational amplifiers. These equations are utilized to the analysis and synthesis of active - R filters. Good correlation has been observed between predicted and experimental results.

광섬유 지연선로를 이용한 3-비트 광 실시간 지연선로

이백송,신종덕,김부균 崇實大學校 生産技術硏究所 2002 論文集 Vol.32 No.-

In this paper, we propose an optical true time-delay (TTD) feeder system for phased array antennas (PAAs), which consists of a fixed wavelength laser diode, 2×2 MEMS (Micro-Electro-Mechanical System) switches, and fiber delay lines. A 3-bit TTD for 10 ㎓ PAAs with two antenna elements has been implemented and the time delay difference between the antenna elements has been measured for all the possible beam radiation angles according to the combination of MEMS switches setup. Experimental results are in excellent agreement with calculated ones.

실시간 지연선로를 사용한 위상배열 안테나의 지향성 향상에 관한 연구

배덕희,신종덕,김부균 崇實大學校 生産技術硏究所 2002 論文集 Vol.32 No.-

In this paper, We proposed a novel true time-delay(TTD) for phased array antennas(PAAs) using fiber Bragg gratings(FBGs) and metal films deposited on fiber ends. This structure requires fewer number of FBGs and provides relaxed wavelengh selection than the previous TTDs using only FBGs. A 10 ㎓ PAA system with the proposed TTD has been fabricated to show that the measured time delay difference for each steering angle is less than ±1ps compared with the calculated one. Dolph-Tschebyscheff array method was used to improve the directivity for a 10 ㎓, four element PAA incorporating the proposed TTD. Simulation results showed that the ratio of the main lobe to the first side lode for the case of Dolph-Tschebyscheff array is 16.2 dB greater than that for the case of uniform array.

The Asymmetry of the Reflectivity Spectrum of Bragg Reflector Structures

Boo Gyoun Kim 대한전자공학회 1990 KITE JOURNAL OF ELECTRONICS ENGINEERING Vol.1 No.2

Spectral Response of Linearly-Chirped Grating Filter

Kim,Boo Gyoun 崇田大學校 1982 論文集 Vol.12 No.1

| Periodic structures that vary spatially in period are analyzed using a WKB method to determine their response as Bragg reflectors or filters. Analytic solutions are expressed in terms of elementary functions. It is shown that the analytic solutions agree well with the numerical solutions based upon the Ricatti equation. Chirped gratings fabricated on thin-film optical waveguides have been used in a variety applications such as optical filters, output couplers and wavelength demultiplexers (3). Kogelnik reduced the pair of coupled-wave eqs. to a single Ricatti eq., which he solved by the fourth order Runge-Kutta method(1). Streifer and coworkers used perturbation technique to determine filter response of nonuniform almost-periodic Bragg reflectors(2). In this letter, we present the analytic solutions of the spectral reflectivity when the gradual variation of the grating period is small. Using WKB method, analytic solutions are expressed in terms of elementary functions. Consider a linearly chirped grating whose period Λ(z) varies as ?? The coupled-wave eqs. become dA/dz=-iKB exp[i( 2δz + ½αz²)] (1a) dB/dz=+iKA exp[-i( 2δz + ½αz²)] , (1b) where A and B are complex amplitudes of the incident and reflected modes under consideration. δ is a phase mismatch factor and is defined by ??, where β is the propagation constant of the uncorrugated waveguide mode. Without loss of generality we normalize A(z) to satisfy A(-1)=1(2a) and since B(z) is generated within the periodic structure B(1)=0(2b). The reflectivity of filter R = |B(-ι)/A(-ι)²| = |B(-ι) ²| (3) gives the filter response for each incident wavelength. To solve Eqs. (1a) and (1b), we eliminate B to obtain d²A/dz²- i(2δ+dz)dA/dz - K²A=0 (4) By means of the substitution A(z)=exp(i¼αz²+ iδz)μ(z) (5) eq. (4) is transformed into Eq. (6) d²μ/dz²- f(z)μ = 0 (6) where f(z) = S²- i½α - δαz - ¼α²z²and s²=K²-δ². (7) Under the assumption S²》α| ι+ ¼αι²+ i½| , ?? = S - α(δz+i½)/2S + 0(α²) (8) Four our convenience, let the length of our grating L=21. With the boundary conditions (2a) and (2b), under the assumption S²》α|δι= ¼αι²+ i½| , we obtain the analytic solution ?? (9) Under the assumption D²》α|δι= ¼αι²+ i½| we obtain the analytic solution ?? (10) where D²= δ²- K². (11) Eqs. (9) and (10) show that the filter response is symmetric for δ=0 and the same at the input and output port of the filter. In order to show the validity of the analytic solutions, they are compared with the numerical solutions based upon Ricatti eq.. The numerical solutions are solved by fourth order Runge-Kutta method. When it is run with an allowed error of ??, ?? per numerical step, the results always agree to four significant digits. We plot R=ρ(-L/2) p*(L/2) as a function of δL for αL²=0.8, 1.6, 2.8 when ΚL=6.0. The analytic and numerical solutions are in good agreement except for narrow region near the turning point. As αL²increases, the difference between analytic and numerical solutions increases near the turning point. This increase occurs because we neglect the terms proportional to α²in deriving analytic solutions. The analytic solutions serve to indicate effects of slight chirping without resorting to lengthy and expensive numerical calculations and to reduce the computer time by a few tenth at least. In conclusion, we have developed analytic solutions for the filter response of near periodic structures. The analytic and numerical solutions are in good agreement except for narrow region near the turning point.

Comparison of the results by the Matrix method and the Coupled wave method in analyzing Bragg Reflector structures

Boo Gyoun Kim 대한전자공학회 1991 KITE JOURNAL OF ELECTRONICS ENGINEERING Vol.2 No.2

기판의 크기가 마이크로스트립 패치 안테나의 방사특성에 미치는 효과

김상우(SangWoo Kim),김태영(Tae-Young Kim),김부균(Boo-Gyoun Kim),신종덕(Jong-Dug Shin),김세윤(Se-Yun Kim) 대한전자공학회 2007 電子工學會論文誌-TC (Telecommunications) Vol.44 No.11

기판의 크기가 마이크로스트립 패치 안테나의 방사특성에 미치는 영향에 대하여 알아보았다. 기판의 폭을 고정시키고 길이를 변화시킬 때 방사패턴 특성이 크게 변화하나 기판 길이를 고정시키고 폭을 변화시킬 때는 방사패턴 특성이 거의 변화하지 않음을 볼 수 있었다. 기판 크기에 따른 방사패턴 특성의 변화는 기판 두께가 두꺼울 때가 작을 때보다 더 큼을 볼 수 있었다. Effect of finite substrate plane on the radiation characteristics of microstrip patch antennas is investigated. The variation of the characteristics of the radiation pattern due to the change of the length of a substrate is much larger than that due to the change of the width of a substrate. As the thickness of a substrate increases, the characteristics of the radiation pattern reveal more sensitive for the change of the size of a substrate.

다단계 AAM 기반 얼굴 윤곽 추출

김대환(Daehwan Kim),김재민(Jaemin Kim),조성원(Seongwon Cho),이기성(Keeseong Lee),장용석(Yongsuk Jang),정선태(Sun-Tae Chung),김부균(Boo-Gyoun Kim) 한국멀티미디어학회 2006 한국멀티미디어학회 학술발표논문집 Vol.2006 No.2

AAM (active appearance model)를 이용한 얼굴 윤곽 추출 방법은 매우 효과적이지만, 초기 값에 매우 민감하다. 초기값이 전역최소값(global optimum)과 많은 차이가 있는 경우에 AAM 기반 알고리즘은 국부최소값(local optimum)으로 수렴할 가능성이 높다. 본 논문은 얼굴 내부의 주요 부위만 포함하는 영역의 특징은 상대적으로 변화가 적고 또한 내부 특징을 AAM으로 찾는 경우 배경 화면의 영향을 적게 받는다는 점에 착안하여 다단계로 얼굴의 특징을 찾는다. 이를 위하여 얼굴 내부 영역만으로 구성된 AAM 모델과 얼굴 전체 영역으로 구성된 AAM 모델을 각각 형성한다. 우선 내부 얼굴 AAM 모델을 이용하여 내부 얼굴 특징 변수와 얼굴의 자세 변수를 추정하고, 추정된 변수 값들을 이용하여 전체 얼굴 특징 변수와 자세 변수 값을 예측한다. 예측방법은 최소자승법(least square method)을 적용한다. 예측된 변수 값들을 전체 얼굴 AAM 모델 변수의 초기값으로 사용하여 전체 얼굴 특징 변수와 자세 변수를 찾는다.

A Linear Phased Array Antenna Composed of Inductive Loaded Patch Antennas

Tae-Young Kim,Young-Min Yoon,Gun-Su Kim,Boo-Gyoun Kim IEEE 2011 IEEE antennas and wireless propagation letters Vol.10 No.-

<P>Radiation characteristics of an inductive loaded patch phased array antenna (PAA) are compared to those of a conventional patch PAA. The performance of an inductive loaded patch PAA is significantly improved compared to that of a conventional patch PAA because the mutual coupling between the adjacent inductive loaded patch antennas is very small compared to that between the adjacent conventional patch antennas. The radiation characteristics of an inductive loaded patch PAA show superior performance such as low variations of the main beam gain and sidelobe level over the variation of the main beam direction.</P>

점진적 AAM 기반 얼굴 윤곽 추출

김대환(Daehwan Kim),김재민(Jaemin Kim),조성원(Seongwon Cho),이기성(Keeseong Lee),장용석(Yongsuk Jang),정선태(Sun-Tae Chung),김부균(Boo-Gyoun Kim) 대한전자공학회 2006 대한전자공학회 학술대회 Vol.2006 No.11

AAM has been successfully applied to face alignment, but its performance is very sensitive to initial value. In case the initial value has some distance from the global optimum value, there exists a pretty good possibility that AAM-based face alignment will converge to a local minimum. In this paper, we propose a progressive AAM-based face alignment algorithm which first find the feature parameters fitting the inner facial feature points of the face and later locate the feature points of the whole face using the first information. This proposed progressive AAM-based face alignment algorithm utilizes the fact that the feature points in the inner part of the face have less varied and less affected by the background surrounding the face than those in the outer part (like the chin contour). The proposed algorithm needs to build two AAM models; firstly the inner AAM model built from only the inner facial feature points of the faces, and secondly the whole AAM model built from the whole facial feature points of the faces. The fitting stage of the algorithm will be processed progressively in two phases. In the first phase, the proposed algorithm will find the feature parameters for the inner facial feature points of the new face, and then in the second phase it will locate the facial feature points of the whole face using the initial value estimated from the inner feature parameters obtained in the first phase. The estimation is based on the least square method. The effectiveness of our proposed algorithm is tested through experiments.