• 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 Journal of the Acoustical Society of Korea
• /
• v.10 no.4
• /
• pp.38-47
• /
• 1991

The frequency variation among speakers is one of problems in the speech recognition. This paper applies fuzzy theory to solve the variation problem of frequency features. Reference patterns are expressed by fuzzified patterns which are produced by the peak frequency and the peak energy extracted from codebooks which are generated from training words uttered by several speakers, as they should include common features of speech signals. Words are recognized by fuzzy inference which uses the certainty factor between the reference patterns and the test fuzzified patterns which are produced by the peak frequency and the peak energy extracted from the power spectrum of input speech signals. Practically, in computing the certainty factor, to reduce memory capacity and computation requirements we propose a new equation which calculates the improved certainty factor using only the difference between two fuzzy values. As a result of experiments to test this word recognition method by fuzzy interence with Korean digits, it is shown that this word recognition method using the new equation presented in this paper, can solve the variation problem of frequency features and that the memory capacity and computation requirements are reduced.

• Journal of Communications and Networks
• /
• v.13 no.6
• /
• pp.655-663
• /
• 2011

Diffuse optical wireless (DOW) systems have the advantage that they do not require point-to-point siting so one transmitter can communicate with several receivers. In this paper, we investigate multiple access scheduling methods for downlink orthogonal frequency division multiplexing (OFDM) in diffuse optical wireless networks. Unlike the radio frequency (RF) channel, the DOW channel has low-pass filter characteristics and so requires different scheduling methods than those developed for the RF channel. Multi-user diversity orthogonal frequency division multiple access (OFDMA) systems nominate a cluster of subcarriers with the largest signal-to-noise-ratio for transmission. However, in a DOW channel, most users would choose the lowest frequency clusters of subcarriers. To remedy this problem, we make two proposals. The first is to use a variable cluster size across the subcarriers; the lower frequency clusters will have fewer subcarriers while the higher frequency clusters will have more subcarriers. This will equalize the capacity of the clusters. The second proposal is to randomize a user's cluster selection from a group of clusters satisfying a minimum threshold. Through simulation it is shown that combining these strategies can increase the throughput while ensuring a fair distribution of the available spectrum.

• Computational Structural Engineering
• /
• v.10 no.4
• /
• pp.185-193
• /
• 1997

A major problem of frequency analysis in the field of low-frequencies such as building or construction vibration is the way of signal processing which is appropriate to obtain included frequency content from the finite process to be measured. Therefore, it is the aim of the investigation reported herein to develop the signal processing algorithm which is analyzed without losing the reliability of the measurements in low-frequency domain. To accomplish the research objective, it was analyzed the problems on the way of signal processing in low-frequency domain, and compared the response characteristics of FFT with those of MEM (Maximum Entropy Method) about the low-frequency of vibration. This evaluation of the response characteristics is used in determining appropriate signal processing algorithm into the low-frequency domain.

• Journal of Power Electronics
• /
• v.10 no.6
• /
• pp.777-783
• /
• 2010

In order to drive gas discharge lamps, DC-AC converters with a LCC resonant tank, whose output voltage is adjusted by a variable frequency control are frequently used. However, when they are activated by varying the operating frequency, converters are frequently damaged by unstable operation, due to the rising and falling of the operating frequency near the resonant frequency. To solve this problem, a simple protection circuit for the power supply of a gas discharge lamp is proposed in this paper. This circuit senses the primary current of the main transformer. Using this protection circuit, the operating frequency of the lamp driving inverter system is kept close to and on the right side of the resonant frequency and the inverter is always operated in the ZVS condition. The resulting stable variable frequency operation allows various gas discharge lamps to be tested without the risk of damaging the main switches, because the protection circuit can protect the power MOSFETs of bridge converters from abnormal conditions. The validity and effectiveness of the proposed protection circuit are verified through the experimental results.

• Journal of the Korean Society of Industry Convergence
• /
• v.21 no.6
• /
• pp.301-307
• /
• 2018

This paper deals with the development of the high frequency thawing machine. The fishery products caught over the world are kept frozen to maintain freshness. These fishery products require thawing before they are sold to customers as food. However, the thawing process can cause freshness reduction, drip coming out, quality deterioration, discharging polluted water, as well as a lot of space and time. The high frequency thawing machine developed to solve this problem has a narrow space, a short thawing time and a small drip. The developed high frequency thawing machine can be used in many fields such as fish processing plant, livestock processing plant. This paper describes the design of the high frequency thawing machine by developing the high frequency generator, development of the controller, and the design of mechanism, and shows the superiority of the high frequency thawing machine by the performance evaluation.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.472-473
• /
• 2021

The frequency-hopping sequence is a type of user-specific code that has been used in various ways in military communication, Bluetooth, and ultra-wideband communications. In a practical communication environment, a frequency-hopping sequence with an alphabet and a length suitable for the number of available frequencies and transmission period is required. Therefore, the design of the frequency-hopping sequence having various parameters is a very important problem in frequency-hopping spread-spectrum communication. To this end, this paper explores a method of synthesizing a sequence of a new length through the synthesis of existing frequency-hopping sequences. As a result, it is possible to present the possibility of frequency-hopping sequences applicable to various environments of communications.

• Journal of The Korean Association For Science Education
• /
• v.28 no.8
• /
• pp.860-869
• /
• 2008

The purpose of this study is to suggest an instructional direction for improving scientific inquiry problem-finding ability of the scientifically-gifted. For this purpose, this study has made an in-depth analysis of the scientific inquiry problems generated by the scientifically-gifted in Problem-Finding Activity in Ill-structured Inquiry Situation (PFAIIS) and Problem-Finding Activity in Well-structured Inquiry Situation (PFAWIS). The results of this study turned out to be as follows: First, most of the problems generated in PFAIIS and PFAWIS could be categorized into seven types (measurement, method, cause, possibility, what, comparison, relationship) according to the inquiry objectives, while the frequency of each type shown in each inquiry objective was a little different. Second, the frequency of scientific concepts stated in inquiry problem was more in PFAWIS than in PFAIIS. But the scientific concepts were shown more diversely in PFAIIS than in PFAWIS. Therefore, results of this study have the following educational implications. First, it is necessary to offer various opportunities of problem-finding activity under ill-structured scientific Inquiry situation. Second, it is needed to emphasize that a new inquiry problem can be found out even during general scientific experiment and frequently to discuss inquiry problems generated during an experiment. Third, it is needed to encourage the scientifically-gifted to generate a scientific inquiry problem based on at least more than seven types.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.12
• /
• pp.2372-2380
• /
• 2015

In this paper, we propose frequency allocation and path selection scheme in underlay cognitive radio (CR) networks using network coding. In the proposed scheme, we choose the path with consideration of network coding and interference temperature in underlay CR networks and propose an optimization problem to maximize the system throughput of secondary users (SUs). Then, we represent the proposed optimization problem as the multi-dimensional multiple-choice knapsack problem and give the theoretical upper bound for the system throughput of SUs by using linear programming. Finally, we compute the system throughput of SUs by using brute-force search (BFS) and link quality first (LQF) scheme in underlay CR networks. Simulation results show that the system throughput of SUs with BFS is higher than that with LQF in underlay CR networks with and without application of network coding, respectively.

• Journal of the military operations research society of Korea
• /
• v.34 no.2
• /
• pp.63-77
• /
• 2008

With the limited frequency spectrum and an increasing demand for cellular communication services, the problem of channel assignment becomes increasingly important. However, finding a conflict free channel assignment with the minimum channel span is NP hard. The innovations are cellular concept, dynamic channel assignment and hierarchical network design. We consider the frequency assignment problem and the base station placement simultaneously. Our model takes the candidate locations emanating from this process and the cost of assigning a frequency, operating and maintaining equipment as an input. Hierarchical network design using genetic algorithm is the first three-tier (Macro, Micro, Pico) model. We increase the reality through applying to Electromagnetic Compatibility Constraints. Computational experiments on 72 problem instances which have $15{\sim}40$ candidate locations demonstrate the computational viability of our procedure. The result of experiments increases the reality and covers 90% of the demand.