• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.103-123
• /
• 1999

Input filtering as a preprocessing method is so much crucial to get good performance in time series forecasting. There are a few preprocessing methods (i.e. ARMA outputs as time domain filters, and Fourier transform or wavelet transform as time-frequency domain filters) for handling time series. Specially, the time-frequency domain filters describe the fractal structure of financial markets better than the time domain filters due to theoretically additional frequency information. Therefore, we, first of all, try to describe and analyze specially some issues on the effectiveness of different filtering methods from viewpoint of the performance of a neural network based forecasting. And then we discuss about neural network model architecture issues, for example, what type of neural network learning architecture is selected for our time series forecasting, and what input size should be applied to a model. In this study an input selection problem is limited to a size selection of the lagged input variables. To solve this problem, we simulate on analyzing and comparing a few neural networks having different model architecture and also use an embedding dimension measure as chaotic time series analysis or nonlinear dynamic analysis to reduce the dimensionality (i.e. the size of time delayed input variables) of the models. Throughout our study, experiments for integration methods of joint time-frequency analysis and neural network techniques are applied to a case study of daily Korean won / U. S dollar exchange returns and finally we suggest an integration framework for future research from our experimental results.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.7
• /
• pp.1001-1007
• /
• 2014

This paper proposes a method to reduce operation time using discrete cosine transform and to improve image quality by the DC gain correction. Conventional filtered back projection (FBP) filtering in the frequency domain using Fourier transform, but the filtering process uses complex number operations. To simplify the filtering process, we propose a filtering process using discrete cosine transform. In addition, the image quality of reconstructed images are improved by correcting DC gain of sinograms. To correct the DC gain, we propose to find an optimum DC weight is defined as the ratio of sinogram DC and optimum DC. Experimental results show that the proposed method gets better performance than the conventional method for phantom and clinical CT images.

• 한국공간정보시스템학회:학술대회논문집
• /
• 2004.05a
• /
• pp.29-36
• /
• 2004

In this paper, a study of damage detection for 2-Dimensional Truss Structures using the parametric projection filter theory is presented. Many researchers are interested in inverse problem and one of solution procedures for inverse problems that are very effective is the approach using the filtering algorithm in conjunction with numerical solution methods. In filtering algorithm, the Kalman filtering algorithm is well known and have been applied to many kind of inverse problems. In this paper, the Parametric projection filtering in conjunction with structural analysis is applied to the identification of damages in 2-D truss structures. The natural frequency and modes of damaged truss model are adopted as the measurement data. The effectiveness of proposed method is verified through the numerical examples.

• Speech Sciences
• /
• v.1
• /
• pp.277-283
• /
• 1997

In this paper, we present a robust recognition algorithm based on the Wiener filtering method as a research tool to develop the Korean Speech recognition system. We especially used Wiener filtering method in cepstrum-domain, because the method in frequency-domain is computationally expensive and complex. Evaluation of the effectiveness of this method has been conducted in speaker-independent isolated Korean digit recognition tasks using discrete HMM speech recognition systems. In these tasks, we used 12th order weighted cepstral as a feature vector and added computer simulated white gaussian noise of different levels to clean speech signals for recognition experiments under noisy conditions. Experimental results show that the presented algorithm can provide an improvement in recognition of as much as from $5\%\;to\;\20\%$ in comparison to spectral subtraction method.

• Journal of Communications and Networks
• /
• v.18 no.1
• /
• pp.8-18
• /
• 2016

In this paper, we propose a number of blind equalization approaches for time-varying andmulti-path channels. The approaches employ cost reference particle filter (CRPF) as the symbol estimator, and additionally employ either least mean squares algorithm, recursive least squares algorithm, or $H{\infty}$ filter (HF) as a channel estimator such that they are jointly employed for the strategy of "Rao-Blackwellization," or equally called "mixture filtering." The novel feature of the proposed approaches is that the blind equalization is performed based on direct channel estimation with unknown noise statistics of the received signals and channel state system while the channel is not directly estimated in the conventional method, and the noise information if known in similar Kalman mixture filtering approach. Simulation results show that the proposed approaches estimate the transmitted symbols and time-varying channel very effectively, and outperform the previously proposed approach which requires the noise information in its application.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.4
• /
• pp.316-320
• /
• 2015

In this paper, the pole filtering concept is applied to the Mel-frequency cepstral coefficient (MFCC) feature vectors in the conventional cepstral mean normalization (CMN) and cepstral mean and variance normalization (CMVN) frameworks. Additionally, performance of the cepstral mean and scale normalization (CMSN), which uses scale normalization instead of variance normalization, is evaluated in speech recognition experiments in noisy environments. Because CMN and CMVN are usually performed on a per-utterance basis, in case of short utterance, they have a problem that reliable estimation of the mean and variance is not guaranteed. However, by applying the pole filtering and scale normalization techniques to the feature normalization process, this problem can be relieved. Experimental results using Aurora 2 database (DB) show that feature normalization method combining the pole-filtering and scale normalization yields the best improvements.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.12
• /
• pp.1113-1119
• /
• 2000

To operate a process plant safely and economically, process monitoring is very important. Process monitoring is the task to identify the state of the system from sensor data. Process monitoring includes data acquisition, regulatory control, data reconciliation, fault detection, etc. This research focuses on the data recon-ciliation using scale-space filtering and fault detection using functional-link associative neural networks. Scale-space filtering is a multi-resolution signal analysis method. Scale-space filtering can extract highest frequency factors(noise) effectively. But scale-space filtering has too large calculation costs and end effect problems. This research reduces the calculation cost of scale-space filtering by applying the minimum limit to the gaussian kernel. And the end-effect that occurs at the end of the signal of the scale-space filtering is overcome by using extrapolation related with the clustering change detection method. Nonlinear principal component analysis methods using neural network have been reviewed and the separately expanded functional-link associative neural network is proposed for chemical process monitoring. The separately expanded functional-link associative neural network has better learning capabilities, generalization abilities and short learning time than the exiting-neural networks. Separately expanded functional-link associative neural network can express a statistical model similar to real process by expanding the input data separately. Combining the proposed methods-modified scale-space filtering and fault detection method using the separately expanded functional-link associative neural network-a process monitoring system is proposed in this research. the usefulness of the proposed method is proven by its application a boiler water supply unit.

• Journal of Ocean Engineering and Technology
• /
• v.2 no.2
• /
• pp.37-45
• /
• 1988

A dynamically positioned vessel must be capable of maintaining a specified position and direction by controlling the thruster devices. The motions of a vessel are often assuned to tne sum of low frequency(LF)motions and high frequency(HF)motions. The former is mainly due to wind, current and second order wave forces, while the latter is mainly due to first order wave forces. In order to avoid the high frequency thruser modulation, the control system must include filters to estimate the low frequency motions from the measured motion signals, This paper presents a control system based on Kalman filtering technique and optimal control tyeory. Using the combined kalmam filter, LF motion estimates and HF ones are achieved from the motion measurement of the vessel. The estimated low frequency motions are used as inputs to the dynamic positioning system. The thruster modulation is minimized using the optimal control theory; Linear Quadratic Gaussian(LQG)controller. The performances of the Kalman filter and the dynamic positioned vessel are investigated by computer simulation.

• The Journal of the Korea Contents Association
• /
• v.8 no.1
• /
• pp.282-289
• /
• 2008

Although there were some technological developments in improving the collaborative filtering, they have yet to fully reflect the actual relation of the items. In this paper, we propose the recommendation system using associative web document classification by word frequency and ${\alpha}$-cut to address the short comings of the collaborative filtering. The proposed method extracts words from web documents through the morpheme analysis and accumulates the weight of term frequency. It makes associative rules and applies the weight of term frequency to its confidence by using Apriori algorithm. And it calculates the similarity among the words using the hypergraph partition. Lastly, it classifies related web document by using ${\alpha}$-cut and calculates similarity by using adjusted cosine similarity. The results show that the proposed method significantly outperforms the existing methods.

• Journal of Biosystems Engineering
• /
• v.14 no.4
• /
• pp.229-241
• /
• 1989

The purposes of this study were to develop an electronic-hydraulic draft control system for tractor implements, to investigate the control performance of the system and the possibility of adaptation to the conventional tractor. Experiments were carried out to investigate the responses of the system to the step and sinusoidal inputs in draft control. The effects of control mode, hydraulic flow rate, reference deadband, and proportional constant on control performance of the system were investigated. Moreover, the effects of filtering signals from draft sensor were also investigated. The following conclusions were derived from the study; 1. In draft control, there were hunting problems in controlling the implement without filtering the draft signals. Filtering was performed by a control program of electronic controller and the control performance and stability of the system were improved significantly. 2. For the draft control system operated on on-off control mode, draft was controlled within ${\pm}27-{\pm}55kg_f$ to the reference draft when the hydraulic flow rates were 5-15 l/min. For the draft control system operated on PWM control, draft was controlled within ${\pm}27kg_f$ to the reference draft regardless of hydraulic flow rates. 3. In the frequency responses of the draft control system, control performance on PWM control mode was not better than on on-off control mode because of characteristics of hydraulic valve and drafe sensor. As the hydraulic flow rates increased for the system operated on on-off control mode, the corner frequency of amplitude attenuation increased, but the corner frequency of phase-angle change remained nearly the same. But, the system was unstable beyond the frequency of 3.1 rad/s. 4. The electronic-hydraulic hitch control system developed in this study showed superior control performance, stability and convenience compared to conventional mechanical-hydraulic hitch control system. It is considered to be a superior replacement for the conventional mechanical-hydraulic hitch control system.