• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.1 no.3
• /
• pp.69-76
• /
• 2008

This paper concerns Fuzzy Radial Basis Function Neural Network (FRBFNN) and automatic rule generation of extraction of the FRBFNN by means of mountain clustering. In the proposed network, the membership functions of the premise part of fuzzy rules do not assume any explicit functional forms such as Gaussian, ellipsoidal, triangular, etc., so its resulting fitness values (degree of membership) directly rely on the computation of the relevant distance between data points. Also, we consider high-order polynomial as the consequent part of fuzzy rules which represent input-output characteristic of sup-space. The number of clusters and the centers of clusters are automatically generated by using mountain clustering method based on the density of data. The centers of cluster which are obtained by using mountain clustering are used to determine a degree of membership and weighted least square estimator (WLSE) is adopted to estimate the coefficients of the consequent polynomial of fuzzy rules. The effectiveness of the proposed model have been investigated and analyzed in detail for the representative nonlinear function.

• Proceedings of the KIEE Conference
• /
• 2007.04c
• /
• pp.29-33
• /
• 2007

In this paper, time-varying parameter of passive telemetry RF sensor system is estimated using RLS(Rescursive $\leq$* Square) algorithm. In order to overcome the problems such as power limits and complication that general RF sensor system including IC chip has, the principle of inductive coupling is applied to model sensor system The model parameter is rearranged for applying RLS algorithm based on mathematical model to the derived model using inductive coupling principle. Time variant parameter of rearranged model is estimated using forgetting factor, and in case measured data is contaminated by noise and modelling error, the performance of RLS algorithm characterized by the convergence of squared error sum is verified by simulation.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.5
• /
• pp.833-842
• /
• 2017

In this study, the design methodology of radial basis function neural networks based on incremental K-means clustering is introduced for learning and processing the big data. If there is a lot of dataset to be trained, general clustering may not learn dataset due to the lack of memory capacity. However, the on-line processing of big data could be effectively realized through the parameters operation of recursive least square estimation as well as the sequential operation of incremental clustering algorithm. Radial basis function neural networks consist of condition part, conclusion part and aggregation part. In the condition part, incremental K-means clustering algorithms is used tweights of the conclusion part are given as linear function and parameters are calculated using recursive least squareo get the center points of data and find the fitness using gaussian function as the activation function. Connection s estimation. In the aggregation part, a final output is obtained by center of gravity method. Using machine learning data, performance index are shown and compared with other models. Also, the performance of the incremental K-means clustering based-RBFNNs is carried out by using PSO. This study demonstrates that the proposed model shows the superiority of algorithmic design from the viewpoint of on-line processing for big data.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.25 no.9
• /
• pp.952-958
• /
• 2014

In this paper, we propose a parametric sparse recovery algorithm(SRA) applied to a radar signal model, based on the compressive sensing(CS), for the ISAR(Inverse Synthetic Aperture Radar) image reconstruction from an incomplete radar-cross-section(RCS) data and for the estimation of rotation rate of a target. As the SRA, the iteratively-reweighted-least-square(IRLS) is combined with the radar signal model including chirp components with unknown chirp rate in the cross-range direction. In addition, the particle swarm optimization(PSO) technique is considered for searching correct parameters related to the rotation rate. Therefore, the parametric SRA based on the IRLS can reconstruct ISAR image and estimate the rotation rate of a target efficiently, although there exists missing data in observed RCS data samples. The performance of the proposed method in terms of image entropy is also compared with that of the traditional interpolation methods for the incomplete RCS data.

• Journal of Korean Institute of Industrial Engineers
• /
• v.42 no.4
• /
• pp.256-262
• /
• 2016

The Weibull is widely used in reliability analysis, and several studies have attempted to improve estimation of the distribution's parameters. least squares estimation (LSE) or Maximum likelihood estimation (MLE) are often used to estimate distribution parameters. However, it has been proven that Bayesian methods are more suitable for small sample sizes than LSE and MLE. In this work, the Weibull parameter estimation accuracy of LSE, MLE, and Bayesian method are compared for sample sets with 3 to 30 data points. The Bayesian method was most accurate for sample sizes under 25, and the accuracy of the Bayesian method was similar to LSE and MLE as the sample size increased.

• International Journal of Ocean System Engineering
• /
• v.1 no.2
• /
• pp.60-67
• /
• 2011

This paper presents an in-situ correction method to compensate for the position error of an autonomous underwater vehicle (AUV) near the sea floor. AUVs generally have an inertial navigation system assisted with auxiliary navigational sensors. Since the inertial navigation system shows drift in position without the bottom reflection of a Doppler velocity log, external acoustic positioning systems, such as an ultra short baseline (USBL), are needed to set the position without surfacing the AUV. The main concept of the correction method is as follows: when the AUV arrives near the sea floor, the vehicle moves around horizontally in a circular mode, while the USBL transceiver installed on a surface vessel measures the AUV's position. After acquiring one data set, a least-square curve fitting method is adopted to find the center of the AUV's circular motion, which is transferred to the AUV via an acoustic telemetry modem (ATM). The proposed method is robust for the outlier of USBL, and it is independent of the time delay for the data transfer of the USBL position with the ATM. The proposed method also reduces the intrinsic position error of the USBL, and is applicable to the in-situ calibration as well as the initialization of the AUVs' position. Monte Carlo simulation was conducted to verify the effectiveness of the method.

• Communications for Statistical Applications and Methods
• /
• v.13 no.3
• /
• pp.567-576
• /
• 2006

In this paper, we applied the multiblock dimension reduction methods to the classification of tumor based on microarray gene expressions data. This procedure involves clustering selected genes, multiblock dimension reduction and classification using linear discrimination analysis and quadratic discrimination analysis.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1999.10a
• /
• pp.275-280
• /
• 1999

This study is object to program development of fatigue test for universal testing machine. Fatigue program is consist of test simulation, data analysis and print report by control fatigue testing program which expansively applies tension-compression tests with using oil pressure mechanism by Visual Basic software running under windows 98.

• The Korean Journal of Applied Statistics
• /
• v.29 no.6
• /
• pp.1117-1128
• /
• 2016

In genetic association studies with high-dimensional genomic data, regularization procedures based on penalized likelihood are often applied to identify genes or genetic regions associated with diseases or traits. A network-based regularization procedure can utilize biological network information (such as genetic pathways and signaling pathways in genetic association studies) with an outstanding selection performance over other regularization procedures such as lasso and elastic-net. However, network-based regularization has a limitation because cannot be applied to high-dimension genomic data with a group structure. In this article, we propose to combine data dimension reduction techniques such as principal component analysis and a partial least square into network-based regularization for the analysis of high-dimensional genomic data with a group structure. The selection performance of the proposed method was evaluated by extensive simulation studies. The proposed method was also applied to real DNA methylation data generated from Illumina Innium HumanMethylation27K BeadChip, where methylation beta values of around 20,000 CpG sites over 12,770 genes were compared between 123 ovarian cancer patients and 152 healthy controls. This analysis was also able to indicate a few cancer-related genes.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.16 no.6
• /
• pp.33-38
• /
• 2008

Most structures of vehicle are composed of many substructures connected to one another by various types of mechanical joints. In vehicle engineering, it is important to study these jointed structures under random frequency vibration for the evaluations of fatigue life and stress concentration exactly. It is rarely obtained the accurate load history of specified positions in a jointed structure because of the errors such as modeling, measurement, and etc. In the beginning of design, exact load data are actually necessary for the fatigue strength and life analysis to minimize the cost and time of designing. In this paper, the hybrid method of practical dynamic load determination is developed by the combination of the principal stresses from F. E. Analysis and test of a jointed structure. Least square pseudo inverse matrix is adopted to obtain an inverse matrix of analyzed stresses matrix. The error minimization method utilizes the inaccurate measured error and the shifting error that the whole data is stiffed over real data. The least square criterion is adopted to avoid these errors. Finally, to verify the proposed system, a heavy-duty bus is analyzed. This measurement and prediction technology can be extended to the different jointed structures.