• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.8
• /
• pp.731-738
• /
• 2011

In this paper, we introduce an advanced architecture of K-Means clustering-based polynomial Radial Basis Function Neural Networks (p-RBFNNs) designed with the aid of SSOA (Space Search Optimization Algorithm) and develop a comprehensive design methodology supporting their construction. In order to design the optimized p-RBFNNs, a center value of each receptive field is determined by running the K-Means clustering algorithm and then the center value and the width of the corresponding receptive field are optimized through SSOA. The connections (weights) of the proposed p-RBFNNs are of functional character and are realized by considering three types of polynomials. In addition, a WLSE (Weighted Least Square Estimation) is used to estimate the coefficients of polynomials (serving as functional connections of the network) of each node from output node. Therefore, a local learning capability and an interpretability of the proposed model are improved. The proposed model is illustrated with the use of nonlinear function, NOx called Machine Learning dataset. A comparative analysis reveals that the proposed model exhibits higher accuracy and superb predictive capability in comparison to some previous models available in the literature.

• Korean Journal of Computational Design and Engineering
• /
• v.5 no.3
• /
• pp.276-284
• /
• 2000

Usual practice of the transformation of a B-spline curve into a set of piecewise polynomial curves in a power form is done by either a knot refinement followed by basis conversions or applying a Taylor expansion on the B-spline curve for each knot span. Presented in this paper is a new algorithm, called a direct expansion algorithm, for the problem. The algorithm first locates the coefficients of all the linear terms that make up the basis functions in a knot span, and then the algorithm directly obtains the power form representation of basis functions by expanding the summation of products of appropriate linear terms. Then, a polynomial segment of a knot span can be easily obtained by the summation of products of the basis functions within the knot span with corresponding control points. Repeating this operation for each knot span, all of the polynomials of the B-spline curve can be transformed into a power form. The algorithm has been applied to both static and dynamic curves. It turns out that the proposed algorithm outperforms the existing algorithms for the conversion for both types of curves. Especially, the proposed algorithm shows significantly fast performance for the dynamic curves.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.1
• /
• pp.135-142
• /
• 2012

In this paper, we proposed a new architecture called radial basis function-based polynomial neural networks classifier that consists of heterogeneous neural networks such as radial basis function neural networks and polynomial neural networks. The underlying architecture of the proposed model equals to polynomial neural networks(PNNs) while polynomial neurons in PNNs are composed of Fuzzy-c means-based radial basis function neural networks(FCM-based RBFNNs) instead of the conventional polynomial function. We consider PNNs to find the optimal local models and use RBFNNs to cover the high dimensionality problems. Also, in the hidden layer of RBFNNs, FCM algorithm is used to produce some clusters based on the similarity of given dataset. The proposed model depends on some parameters such as the number of input variables in PNNs, the number of clusters and fuzzification coefficient in FCM and polynomial type in RBFNNs. A multiobjective particle swarm optimization using crowding distance (MoPSO-CD) is exploited in order to carry out both structural and parametric optimization of the proposed networks. MoPSO is introduced for not only the performance of model but also complexity and interpretability. The usefulness of the proposed model as a classifier is evaluated with the aid of some benchmark datasets such as iris and liver.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.57-61
• /
• 2016

교각 주위에서의 국부세굴은 교각을 지나는 유체의 복잡한 흐름에 의해 발생한다. 이를 해석하기 위하여 많은 난류모형을 이용한 실내실험 및 수치실험을 수행하였으나 발생하는 와류를 하천 규모에서 전부 계산하기는 매우 어려운 문제다. 따라서 국부세굴 관련으로 최대 관심사인 최대 세굴심은 인공지능 기술에 근거한 다양한 기법을 적용해 계산하여 예측하기도 한다. 본 연구에서는 기계학습 분야 중 하나인 서포트 벡터 머신 (Support Vector Machines)을 이용하여 교각주위 국부세굴을 예측하였다. SVM은 본래 초평면을 이용하여 데이터를 분류시키는 기법이나 Vapnik(1995)이 제안한 ${\varepsilon}$ 서포트 벡터 회귀 (${\varepsilon}$-support vector regression)방법을 통해 회귀분석에도 활용할 수 있게 되었다. 학습을 위해 Charbert and Engeldinger (1956), Shen et al. (1969), Jain and Fischer (1979), 그리고 Dey et al. (1995)의 실험 자료를 이용하였고 검증을 위해 Yanmaz and Altinbilek (1991)의 실험 자료를 이용하였다. 커널함수로는 다항식 함수와 방사 기저 함수를 이용하였고 각 계수는 적합한 값을 찾기 위해 시행착오법을 사용하였다. 민감도 분석을 통해 각 계수들 중 ${\varepsilon}$의 변화가 결과에 가장 민감하게 변화를 일으키는 것을 확인하였고 검증 결과 SVM가 충분히 국부세굴을 잘 예측하는 것을 확인하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.1
• /
• pp.1-6
• /
• 2014

In this study, we propose optimized RBFNNs based on night vision face recognition simulator with the aid of $(2D)^2$ PCA algorithm. It is difficult to obtain the night image for performing face recognition due to low brightness in case of image acquired through CCD camera at night. For this reason, a night vision camera is used to get images at night. Ada-Boost algorithm is also used for the detection of face images on both face and non-face image area. And the minimization of distortion phenomenon of the images is carried out by using the histogram equalization. These high-dimensional images are reduced to low-dimensional images by using $(2D)^2$ PCA algorithm. Face recognition is performed through polynomial-based RBFNNs classifier, and the essential design parameters of the classifiers are optimized by means of Differential Evolution(DE). The performance evaluation of the optimized RBFNNs based on $(2D)^2$ PCA is carried out with the aid of night vision face recognition system and IC&CI Lab data.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.12 no.2
• /
• pp.45-52
• /
• 2002

In this contributions, we propose a new MSB(most significant bit) algorithm based on AOP(All One Polynomial) and two parallel semi-systolic architectures to computes $AB^2$over finite field $GF(2^m)$. The proposed architectures are based on standard basis and use the property of irreducible AOP(All One Polynomial) which is all coefficients of 1. The proposed parallel semi-systolic architecture(PSM) has the critical path of $D_{AND2^+}D_{XOR2}$ per cell and the latency of m+1. The modified parallel semi-systolic architecture(WPSM) has the critical path of $D_{XOR2}$ per cell and has the same latency with PSM. The proposed two architectures, PSM and MPSM, have a low latency and a small hardware complexity compared to the previous architectures. They can be used as a basic architecture for exponentiation, division, and inversion. Since the proposed architectures have regularity, modularity and concurrency, they are suitable for VLSI implementation. They can be used as a basic architecture for algorithms, such as the Diffie-Hellman key exchange scheme, the Digital Signature Algorithm(DSA), and the ElGamal encryption scheme which are needed exponentiation operation. The application of the algorithms can be used cryptosystem implementation based on elliptic curve.

• Journal of the Korea Convergence Society
• /
• v.12 no.3
• /
• pp.187-196
• /
• 2021

The paper deals with comparative study for characteristics of approximation of design space according to various approximate models and sensitivity analysis using orthogonal array experiments in structure design of active type DSF which was developed for float-over installation of offshore plant. This study aims to propose the orthogonal array experiments based design methodology which is able to efficiently explore an optimum design case and to generate the accurate approximate model. Thickness sizes of main structure member were applied to the design factors, and output responses were considered structure weight and strength performances. Quantitative effects on the output responses for each design factor were evaluated using the orthogonal array experiment. Best design case was also identified to improve the structure design with weight minimization. From the orthogonal array experiment results, various approximate models such as response surface model, Kriging model, Chebyshev orthogonal polynomial model, and radial basis function based neural network model were generated. The experiment results from orthogonal array method were validated by the approximate modeling results. It was found that the radial basis function based neural network model among the approximate models was able to approximate the design space of the active type DSF with the highest accuracy.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.1
• /
• pp.37-43
• /
• 2020

In this paper, we propose an efficient AB² multiplication algorithm using SPB(shifted polynomial basis) over finite fields. Using the feature of the SPB, we split the equation for AB² multiplication into two parts. The two partitioned equations are executable at the same time, and we derive an algorithm that processes them in parallel. Then we propose an efficient semi-systolic AB² multiplier based on the proposed algorithm. The proposed multiplier has less area-time (AT) complexity than related multipliers. In detail, the proposed AB² multiplier saves about 94%, 87%, 86% and 83% of the AT complexity of the multipliers of Wei, Wang-Guo, Kim-Lee, Choi-Lee, respectively. Therefore, the proposed multiplier is suitable for VLSI implementation and can be easily adopted as the basic building block for various applications.

• The KIPS Transactions:PartA
• /
• v.15A no.3
• /
• pp.161-166
• /
• 2008

In this paper, an efficient digit-serial systolic array is proposed for multiplication in finite field GF($2^m$) using the polynomial basis representation. The proposed systolic array is based on the most significant digit first (MSD-first) multiplication algorithm and produces multiplication results at a rate of one every "m/D" clock cycles, where D is the selected digit size. Since the inner structure of the proposed multiplier is tree-type, critical path increases logarithmically proportional to D. Therefore, the computation delay of the proposed architecture is significantly less than previously proposed digit-serial systolic multipliers whose critical path increases proportional to D. Furthermore, since the new architecture has the features of a high regularity, modularity, and unidirectional data flow, it is well suited to VLSI implementation.

• Journal of the Korea Computer Industry Society
• /
• v.6 no.5
• /
• pp.675-686
• /
• 2005

This paper proposes a new method of mathematically modeling and computer simulating television ghosts wherein television signals that have undergone multipath fading are generated without using approximations by considering the attenuation, time delay, phase, and timing jitter between consecutive frames. Conventional methods used polynomial interpolation or complex arithmetic to take into account the ghost phase, but our method uses only real arithmetic by employing the Hilbert transform and also reduces the computation time using the FFT (fast Fourier transform) algorithm. Furthermore, it is also possible to observe the transmit waveforms in both RF and IF ranges. Various ghost patterns generated in software provide for essential data required for the development of ghost canceling algorithms, and are deemed to be very useful in analyzing the constituent blocks of the transmitter and receiver chain in television broadcasting. The development of ghost cancelers needs to be preceded by the task of mathematically modeling ghosts and their extensive computer simulations.