• The KIPS Transactions:PartB
• /
• v.10B no.6
• /
• pp.587-592
• /
• 2003

Many real world control problems have continuous states and actions. When the state space is continuous, the reinforcement learning problems involve very large state space and suffer from memory and time for learning all individual state-action values. These problems need function approximators that reason action about new state from previously experienced states. We introduce Fuzzy Q-Map that is a function approximators for 1 - step Q-learning and is based on fuzzy clustering. Fuzzy Q-Map groups similar states and chooses an action and refers Q value according to membership degree. The centroid and Q value of winner cluster is updated using membership degree and TD(Temporal Difference) error. We applied Fuzzy Q-Map to the mountain car problem and acquired accelerated learning speed.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2000.11a
• /
• pp.357-365
• /
• 2000

In this paper, we proposed a new circuit construction method that reduced the number of CMOS devices of singleton fuzzy controller(SFC) through the proposing a new membership function circuit(MFC) which uses the language variable selecting and the coefficient selecting circuit. According to the range of input values, we can choose the language variables beforehand which will be used in the inference. So we proposed the new MFC which generates the only necessary language variables. Also, we removed all rules of which adapting degree of their antecedents is zero through proposing the coefficient selecting circuit which beforehand selects the coefficients which will influence the inference result. Though this method, we simplified the structure of SFC and reduced the size of hardware. And to solve the problem in the current mode with respect to the restriction of the fan-out number, voltage-input and current-out membership function circuits are constituted of operational transconductance amplifiers. A membership function circuit which includes the language variable selecting circuit, a minimum operation circuit we implemented by current mode CMOS devices. As a result of applying proposed method, total numbers of blocks and devices wave decreased. If the number of variables and antecedents are getting larger, this method is more efficient.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.1
• /
• pp.44-50
• /
• 2007

In this paper generating algorithms of the conventional chaos code sequences and a new chaos code sequence derived from the real and imaginary parts of FFT of one chaos code sequence, and the time-frequency function characteristics of generated code sequences to measure the signal resolution, are considered. And two chaos based QCSK digital communication systems-one with two different chaos code sequences, the other with FFT of one chaos code sequence-are analyzed in AWGN and fading communication channels. After analyzing the correlation functions of such sequences, the delta-like autocorrelation and near zero crosscorrelation functions of them are very suitable for chaos based spread spectrum communication systems, is verified. Through evaluating the performance of two chaos based QCSK systems above using Monte-Carlo simulation, the improvement of performance in the latter QCSK system compare favorably with that of the former system, is shown.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.6
• /
• pp.1269-1274
• /
• 2014

In computer vision research, binarization procedure is one of the most frequently used tools to discriminate target objects from background in grey level binary image. Fuzzy binarization is a reliable technique in environment with high uncertainty such as medical image analysis by setting the threshold as the average of minimum and maximum brightness with triangle type fuzzy membership function. However, this technique is also known as contrast sensitive method thus its discrimination power is not so great when the image has low contrast difference between objects and backgrounds and suffer from information loss as a result. Thus, in this paper, we propose a fuzzy binarization using ART2 algorithm to handle such low contrast image analysis. Proposed ART2 algorithm is applied to determine the medium point of membership function in the fuzzy binarization paradigm. The proposed methods shows low information loss rate in our experiment.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.8 no.4
• /
• pp.403-414
• /
• 1997

The distribution of the space debris caused by any source affects the DBS & FSS wave propagation for satellite communication. In this paper, the normalized expansion coefficients are evaluated with varying the volume distribution density of space debris and the operating frequency, and then the attenuation and phase shift are calculated by using the normalized expansion coefficients. Conclusively, the attenuation and phase shift are mostly affected with the real and imaginary part of the amplitude function, respectively, which represented by electric field component of the Bessel function. The results of this paper adapted for the design of the link budget for satellite communication system and the reconsideration of space environment as the space debris increses by geometric progression.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.32 no.6B
• /
• pp.373-378
• /
• 2012

In this study, a new empirical equation for the transverse dispersion coefficient has been developed based on the theoretical background in river bends. The nonlinear least-square method was applied to determine regression coefficients of the equation. The estimated dispersion coefficients derived by the new equation were compared with observed transverse dispersion coefficients acquired from natural rivers and coefficients calculated by the other existing empirical equations. From a comparison of the existing transverse dispersion equations and the new proposed equation, it appears that the behavior of the existing formula in a relative sense is very much dependent on the friction factor and the river geometry. However, the new proposed equation does not vary widely according to variation of friction factor. Also, it was revealed that the equation proposed in this study becomes an asymptotic curve as the curvature effect increases.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.10
• /
• pp.2819-2830
• /
• 2009

Load bearing structural members in a wide variety of applications accumulate damage over their service life. During experiment much effort and cost is needed for measuring structural safety assessment. The sparseness and errors of measured data have to be considered during the safety estimation of structures. This paper introduces parameter estimation and damage identification algorithm by a system identification using static and dynamic response. The equation error estimator and response error widely used in system identification are based on the minimization of least squared error between measured and calculated responses by a mathematical model of a structure. Since each estimator has a specific form of application in noisy environment and proposes different definitions for these forms. To study the behaviour of the estimators in noisy environment Using Monte Carlo simulation, and a data measured pertubation scheme is adopted to investigate the influence of measurement errors on identification results. The assessment result by static and dynamic response were compared, and the efficiency and applicabilities of the proposed algorithm are demonstrated through simulated static and dynamic responses of a dimensional truss type structures.

• Tunnel and Underground Space
• /
• v.27 no.4
• /
• pp.243-252
• /
• 2017

Since the failure behavior of transversely isotropic rocks is significantly different from that of isotropic rocks, it is necessary to develop a transversely isotropic rock failure function in order to evaluate the stability of rock structures constructed in transversely isotropic rock masses. In this study, a spatial distribution function for strength parameters of transversely isotropic rocks is proposed, which is based on the Cassini oval curve proposed by 17th century astronomer Giovanni Domenico Cassini to model the orbit of the Sun around the Earth. The proposed distribution function consists of two model parameters which could be identified through triaxial compression tests on transversely isotropic rock samples. The original Mohr-Coulomb (M-C) failure function is extended to a three-dimensional transversely isotropic M-C failure function by employing the proposed strength parameter distribution function for the spatial distributions of the friction angle and cohesion. In order to verify the suitability of the transversely isotropic M-C failure function, both the conventional triaxial compression and true triaxial compression tests of transversely isotropic rock samples are simulated. The predicted results from the numerical experiments are consistent with the failure behavior of transversely isotropic rocks observed in the actual laboratory tests. In addition, the simulated result of true triaxial compression tests hints that the dependence of rock strength on intermediate principal stress may be closely related to the distribution of the microstructures included in the rock samples.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.18 no.2
• /
• pp.112-123
• /
• 2006

Double-peak normal distribution function was suggested as the probability density function of the non-tidal components (NTC) data in Korean coastal zone. Frequency distribution analysis of the NTC data was carried out using hourly tidal elevation data of the ten tidal gauging stations, i.e., Incheon, Gunsan, Mokpo, Jeju, Yeosu, Masan, Gadeokdo, Busan, Pohang, and Sokcho which were served through the Internet Homepage by the National Ocean Research Institute. NTC data is defined as the difference between the measured tidal elevation data and the astronomical tidal elevation data using 64 tidal constituents information. Based on the RMS error and R2 value comparison analysis, it was found that this suggested function as the probability density function of the NTC data was found to be more appropriate than the normal distribution function. The parameters of the double-peak function were estimated optimally using Levenberg-Marquardt method which was modified from the Newton method. The standard deviation and skewness coefficient were highly correlated with the non-tidal constants of the tidal gauging stations except Mokpo, Jeju and Sokcho stations.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.9
• /
• pp.1686-1692
• /
• 2015

This paper proposes a modified genetic algorithm that has the same beamforming performance and a fast convergence speed using general genetic algorithm in order to form a beam for the mobile node in a mesh network. The proposed beamforming genetic algorithm selects a part of chromosome a high fitness value in mating process to obtain fast convergence speed, and rest part of chromosome with longer fitness value in order to avoid local solution. Furthermore, the reference beam pattern with Gaussian shape reduces additional convergence speed. Simulation shows that the convergence speed of proposed algorithm improves 20% compared with that of conventional beamforming genetic algorithm.