• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.44 no.5
• /
• pp.416-422
• /
• 2016

This paper presents preliminary performance analysis of a satellite formation flying testbed, which is under development by Astrodynamics and Control Laboratory, Department of Astronomy, Yonsei University. A model reference adaptive controller (MRAC) with a first-order reference model is chosen to enhance the response of reaction wheel system which is subject to uncertainties caused by unmodelled dynamics and measurement noise. In addition, an on-line parameter estimation (OPE) technique based on the least square is combined to eliminate the effect of angular measurement noise by estimating the moment of inertia. Both numerical simulations and hardware experiments with MRAC support the effectiveness and applicability of the adaptive control scheme, which maintains the tracking error below $0.25^{\circ}$ for the entire time span. However, the high frequency control input generated in hardware experiment strongly suggests design modifications to reduce the effect of deadzone.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1994.04a
• /
• pp.16-21
• /
• 1994

For the systems such as ANC(Active Noise Control) systems having auxiliary path after FIR type adaptive filter, Filtered-X LMS algorithm is effective. However behaviors of this algorithm has not been fully understood. The convergence property of this algorithm depends on not only cross correlation matrix between the filtered signals through model and real auxiliary path state solution of weight vector in Filtered-X LMS algorithm is investigated for under-determined case, over-determined case, and nonsingular case. Also, the convergence speed in case of two tone noise is investigated based on the eigenvalue spread of cross correlation matrix.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.1
• /
• pp.436-450
• /
• 2017

Recently, the method based on fractional order partial differential equation has been used in image processing. Usually, the optional order of fractional differentiation is determined by a lot of experiments. In this paper, a denoising model is proposed based on adaptive fractional order anisotropic diffusion. In the proposed model, the complexity of the local image texture is reflected by the local variance, and the order of the fractional differentiation is determined adaptively. In the process of the adaptive fractional order model, the discrete Fourier transform is applied to compute the fractional order difference as well as the dynamic evolution process. Experimental results show that the peak signal-to-noise ratio (PSNR) and structural similarity index measurement (SSIM) of the proposed image denoising algorithm is better than that of other some algorithms. The proposed algorithm not only can keep the detailed image information and edge information, but also obtain a good visual effect.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1103-1108
• /
• 2023

In this paper, we propose a new algorithm for attenuating the background noises in acoustic signal. This algorithm improves the noise attenuation performance by using the FNN(: Full-connected Neural Network) deep learning algorithm instead of the existing adaptive filter after wavelet transform. After wavelet transforming the input signal for each short-time period, noise is removed from a single input audio signal containing noise by using a 1024-1024-512-neuron FNN deep learning model. This transforms the time-domain voice signal into the time-frequency domain so that the noise characteristics are well expressed, and effectively predicts voice in a noisy environment through supervised learning using the conversion parameter of the pure voice signal for the conversion parameter. In order to verify the performance of the noise reduction system proposed in this study, a simulation program using Tensorflow and Keras libraries was written and a simulation was performed. As a result of the experiment, the proposed deep learning algorithm improved Mean Square Error (MSE) by 30% compared to the case of using the existing adaptive filter and by 20% compared to the case of using the STFT(: Short-Time Fourier Transform) transform effect was obtained.

• The Journal of the Korea Contents Association
• /
• v.9 no.12
• /
• pp.10-18
• /
• 2009

Color rendition method is necessary for improving the low contrast images which are achieved by PDA, mobile phone camera or PC camera. There are some methods for color rendition. However, after correcting the color, image quality degradations, such as graying-out, halo-artifact and color noise, may occur. In order to overcome these problems, this paper proposes a retinex-based color rendition method. The proposed method uses the HSV color coordinate system to avoid the graying-out, and the advanced image formation model to reduce the halo-artifact in which the image is divided into three components as the global illumination, the local illumination, and reflectance. The experiment results show that the proposed method yields better performance of color correction over the conveniently method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.8B
• /
• pp.1518-1531
• /
• 1999

The reconstructed images from highly compressed MPEG or H.263 data have noticeable image degradations, such as blocking artifacts near the block boundaries, corner outliers at cross points of blocks, and ringing noise near image edges, because the MPEG or H.263 quantizes the transformed coefficients of 8$\times$8 pixel blocks. A post-processing algorithm has been proposed by authors to reduce quantization effects, such as blocking artifacts, corner outliers, and ringing noise, in MPEG-decompressed images. Our signal-adaptive post-processing algorithm reduces the quantization effects adaptively by using both spatial frequency and temporal information extracted from the compressed data. The blocking artifacts are reduced by one-dimensional (1-D) horizontal and vertical low pass filtering (LPF), and the ringing noise is reduced by two-dimensional (2-D) signal-adaptive filtering (SAF). A comparison study of the subjective quality evaluation using modified single stimulus method (MSSM), the objective quality evaluation (PSNR) and the computation complexity analysis between the signal-adaptive post-processing algorithm and the MPEG-4 VM (Verification Model) post-processing algorithm is performed by computer simulation with several MPEG-4 image sequences. According to the comparison study, the subjective image qualities of both algorithms are similar, whereas the PSNR and the comparison complexity analysis of the signal-adaptive post-processing algorithm shows better performance than the VM post-processing algorithm.

• Journal of Korea Water Resources Association
• /
• v.46 no.10
• /
• pp.969-976
• /
• 2013

Leakage in water distribution system causes social and economic losses by direct water loss into the ground, and additional energy demand for water supply. This research suggests a leak detection model of using adaptive Kalman filtering on real-time data of pipe flow. The proposed model takes into account hourly and daily variations of water demand. In addition, the model's prediction accuracy is improved by automatically calibrating the covariance of noise through innovation sequence. The adaptive Kalman filtering shows more accurate result than the existing Kalman method for virtual sine flow data. Then, the model is applied to data from two real district metered area in JE city. It is expected that the proposed model can be an effective tool for operating water supply system through detecting burst leakage and abnormal water usage.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.1
• /
• pp.93-100
• /
• 2003

Accuracy in maneuvering target tracking using multiple models relies upon the suit-ability of each target motion model to be used. To construct multiple models, the interacting multiple model (IMM) algorithm and the adaptive IMM (AIMM) algorithm require predefined sub-models and predetermined acceleration intervals, respectively, in consideration of the properties of maneuvers. To solve these problems, this paper proposes the GA-based IMM method as an intelligent tracking method for a maneuvering target. In the proposed method, the acceleration input is regarded as an additive process noise, a sub-model is represented as a fuzzy system to compute the time-varying variance of the overall process noise, and, to optimize the employed fuzzy system, the genetic algorithm (GA) is utilized. The simulation results show that the proposed method has a better tracking performance than the AIMM algorithm.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2003.11a
• /
• pp.104-108
• /
• 2003

Finite element (FE) model updating is a procedure to minimize the differences between analytical and experimental results, which can be usually posed as an optimization problem. This paper aims to introduce a hybrid optimization algorithm (GA-SA), which consists of a Genetic algorithm (GA) stage and an Adaptive Simulated Annealing (ASA) stage, to FE model updating for a shrunk shaft. A good agreement of the first four natural frequencies has been achieved obtained from GASA based updated model (FEgasa) and experiment. In order to prove the validity of GA-SA, comparisons of natural frequencies obtained from the initial FE model (FEinit), GA based updated model (FEga) and ASA based updated model (FEasa) are carried out. Simultaneously, the FRF comparisons obtained from different FE models and experiment are also shown. It is concluded that the GA, ASA, GA-SA are powerful optimization techniques which can be successfully applied to FE model updating, the natural frequencies and FRF obtained from all the updated models show much better agreement with experiment than that obtained from FEinit model. However, FEgasa is proved to be the most reasonable FE model, and also FEasa model is better than FEga model.

• ETRI Journal
• /
• v.30 no.6
• /
• pp.865-867
• /
• 2008

A conventional environment adaptation for robust speech recognition is usually conducted using transform-based techniques. Here, we present a discriminative adaptation strategy based on a multi-condition-trained model, and propose a new method to provide universal application to a new environment using the environment's specific conditions. Experimental results show that a speech recognition system adapted using the proposed method works successfully for other conditions as well as for those of the new environment.