• 융합신호처리학회논문지
• /
• 제6권2호
• /
• pp.82-88
• /
• 2005

본 논문은 선형적으로 파라미터화된 시스템에 대한 보상방식을 제안한다. 이 보상기는 전형적인 선형 제어기와 적분형의 적응법칙을 갖는 적응 관측기로 구성되며 이 때 적응법칙은 SG 알고리즘에 근거하여 설계된다. 제안된 보상전략에서는 다른 여러 연구에서 제안된 중간함수 대신에 growth조건, convex조건, attainability조건, 그리고 pseudo gradient 조건을 만족하는 함수들로 적응법칙이 설계된다. 제안된 방식은 추적오차에 대한 점근적 안정도 및 파라미터에 대한 추정오차의 bounded stability를 만족한다. 예제를 통하여 제안된 보상방식의 타당성을 보인다. 그리고 기존의 방식인 Huang의 방법과의 비교를 통해 제안된 방식이 정상상태에서의 파라미터 오차가 더 작아짐을 보인다.

• 한국정보통신학회논문지
• /
• 제9권1호
• /
• pp.152-159
• /
• 2005

본 논문에서는 스텝 크기(step size)를 자동적으로 조절함으로써 빠른 수렴비와 낮은 초과 MSE를 갖는 TS(Tagaki-Sugeno) 퍼지 모델과 ISI에 강하고 위상변화에 둔감한 CMA(constant modulus algorithm)를 접목시킨 새로운 퍼지 확률 기울기(Fuzzy Stochastic Gradient) 알고리즘을 제시하고 비이상적인 전송채널에 의해서 발생한 왜곡을 보상함으로써 수신 측에서 비트 검출 오류를 감소시키기 위하여 국내 지상파 디지털 TV의 표준으로 되어 있는 VSB 방식에 적용 가능한 등화기(equalizer)를 구현하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.1033-1036
• /
• 1999

This paper proposes an efficient fault diagnosis for digital circuits using multilayer neural networks. The efficient learning algorithm is also proposed for the multilayer neural network, which is combined the steepest descent for high-speed optimization and the dynamic tunneling for global optimization. The fault-diagnosis system using the multilayer neural network of the proposed algorithm has been applied to the parity generator circuit. The simulation results shows that the proposed system is higher convergence speed and rate, in comparision with system using the backpropagation algorithm based on the gradient descent.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2022년도 학술발표회
• /
• pp.150-150
• /
• 2022

This study investigates the roles of decomposition methods on high accuracy in daily rainfall prediction from light gradient boosting machine (LightGBM) algorithm. Here, empirical mode decomposition (EMD) and singular spectrum analysis (SSA) methods were considered to decompose and reconstruct input time series into trend terms, fluctuating terms, and noise components. The decomposed time series from EMD and SSA methods were used as input data for LightGBM algorithm in two hybrid models, including empirical mode-based light gradient boosting machine (EMDGBM) and singular spectrum analysis-based light gradient boosting machine (SSAGBM), respectively. A total of four parameters (i.e., temperature, humidity, wind speed, and rainfall) at a daily scale from 2003 to 2017 is used as input data for daily rainfall prediction. As results from statistical performance indicators, it indicates that the SSAGBM model shows a better performance than the EMDGBM model and the original LightGBM algorithm with no decomposition methods. It represents that the accuracy of LightGBM algorithm in rainfall prediction was improved with the SSA method when using multivariate dataset.

• Journal of Communications and Networks
• /
• 제12권6호
• /
• pp.600-604
• /
• 2010

A method of width adaptation in the radial basis function network (RBFN) using stochastic gradient (SG) algorithm is introduced. Using Taylor's expansion of error signal and differentiating the error with respect to the step-size, the optimal time-varying step-size of the width in RBFN is derived. The proposed approach to adjusting widths in RBFN achieves superior learning speed and the steady-state mean square error (MSE) performance in nonlinear channel environment. The proposed method has shown enhanced steady-state MSE performance by more than 3 dB in both nonlinear channel environments. The results confirm that controlling over step-size of the width in RBFN by the proposed algorithm can be an effective approach to enhancement of convergence speed and the steady-state value of MSE.

• 한국연초학회지
• /
• 제20권1호
• /
• pp.87-94
• /
• 1998

After we made the computer source code with mathematical model of Muramatsu et al. that was expressed by the set of simultaneous first-order ordinary differential equations in evaporation-pyrolysis zone of cigarette, we simulated the distribution profiles of temperature and density of flue-cured tobacco. Those equations were solved numerically with the Runge-Kutta-Gill algorithm assuming step size of 0.025mm by Muramatsu at at,, but in this study the advanced algorithm of Runge-Kutta 4th Order assuming step size of 0.0005mm. The initial conditions and physical parameters of Muramatsu et at. were used for solving them. The calculated values corresponded well with results of Muramatsu et al., especially the gradient of the temperature profile increased with smoldering speed and the thickness of the evaporation-pyrolysis zone decreased with increasing of smoldering speed. On the other hand, the temperature gradient decreased with increasing of the effective thermal-conductivity value and the thickness of the evaporation-pyrolysis zone increased with the effective thermal-conductivity value.

• Journal of Mechanical Science and Technology
• /
• 제16권8호
• /
• pp.1033-1038
• /
• 2002

Genetic alsorithm (GA) , compared to the gradient-based optimization, has advantages of convergence to a global optimized solution. The genetic algorithm requires so many number of analyses that may cause high computational cost for genetic search. This paper proposes a personal computer network programming based on TCP/IP protocol and client-server model using socket, to improve processing speed of the genetic algorithm for optimization of composite laminated structures. By distributed processing for the generated population, improvement in processing speed has been obtained. Consequently, usage of network-based genetic algorithm with the faster network communication speed will be a very valuable tool for the discrete optimization of large scale and complex structures requiring high computational cost.

• 전자공학회논문지B
• /
• 제29B권11호
• /
• pp.73-84
• /
• 1992

In this paper, an algorithm for recovering the 3-D shape from a single shaded image is proposed. In the proposed algorithm, by using the relation between the height and surface gradient (p, q), a set of linear equations is derived from the linearized reflectance function. Then the 3-D surface is recovered by employing the conjugate gradient technique. In order to improve the convergence speed of the solution, we also employ the hierarchical basis function and multiresolution images in the algorithm. A method for determining the regularization parameter, which is determined by trial and error in the conventional approach, is also introduced. In addition, the proposed algorithm attempts to recover the 3-D surface without requiring the boundary conditions, making it suitable for a real-time implementation. Simulation results for real image as well as synthetic image are provided to demonstrate the performance of the proposed algorithm.

• 한국통신학회논문지
• /
• 제28권3C호
• /
• pp.232-238
• /
• 2003

본 논문에서는 공간 영역의 제약 정보를 이용한 적응 영상 복원 방식을 제안한다. 공간 영역의 제약정보로는 국부 정보의 평균, 분산 및 최대 값을 이용하였으며, 반복 기법을 이용하여 매 반복 해에서 얻어진 복원 영상으로부터 상기 제약 정보를 설정하게 되고, 위의 제약 정보는 임의의 입력 값에 의해 정의되는 매개 변수와 더불어 복원 영상의 국부 완화 정도를 결정하게 된다. 제안된 방식을 이용하여 복원영상을 얻기 위해 비 적응 복원 방식보다 빠른 수렴속도를 갖게 됨을 알 수 있으며, 국부적으로 제어된 완화 정도를 지닌 복윈 영상을 얻을 수 있었다. 제안된 방식의 성능은 실험을 통해서 확인할 수 있었다.

• 한국통신학회논문지
• /
• 제25권7A호
• /
• pp.978-985
• /
• 2000

블라인드 등화기는 학습신호를 이용하지 않고 저송된 데이터의 알려진 특성을 이용해 신호를 복원하며 일반적으로 가장 많이 이용되는 알고리즘은 구현이 간단한 Steepest Gradient Descent 계열의 알고리즘으로서 CMA나 Sato 알고리즘이 여기에 속한다. 본 논문에선, CMA 및 Normalized CMA (NCMA)의 장점과 이중모드 위상복원 알고리즘의 장점을 결합하는 이중모드 NCMA 알고리즘을 제안하고 QAM 시스템에 적용한 컴퓨터 시뮬레이션을 수행하여 제안한 알고리즘이 CMA와 이중모드 CMA 보다 더 빠른 수렴속도와 더 적은 정상상태 잔류에서 특성을 가짐을 확인하다.