• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.43 no.6
• /
• pp.1027-1035
• /
• 1994

This paper aims to make an artificial arm control strategy. For this, we propose a new feature extraction method and design artificial neural network for the functional classification of myoelectric signal(MES). We first transform the two channel myoelectric signals (MES) for biceps and triceps into frequency domain using fast Fourier transform (FFT). And features were obtained by comparing the magnitudes of ensemble spectrum data and used as inputs to the three-layer neural network for the learning. By changing the number of units in hidden layer of neural network we observed the improvement of classification performance. To observe the effeciency of the proposed scheme we performed experiments for classification of six arm functions to the three subjects. And we obtained on average 94[%] the ratio of classification.

• IEIE Transactions on Smart Processing and Computing
• /
• v.3 no.3
• /
• pp.135-141
• /
• 2014

The inherent feature of the highly efficient spectrum usage has made Orthogonal Frequency Division Multiplexing (OFDM) preferable for Communication Standards. This study evaluated the performance of a Least Square (LS) estimator for a comb-type pilot insertion scheme over a fast fading Rayleigh channel. A High Peak-to-Average Power Ratio (PAPR) is one of the major downsides of the OFDM. The effects of an increase in the number of subcarriers on PAPR and the performance of the LS Estimator were studied. Increasing the number of subcarriers while keeping the pilots overhead constant resulted in improved performance of the LS estimator but the PAPR increased with increasing number of subcarriers. Therefore some trade-off between the number of subcarriers and the performance of the OFDM system is needed. The Mean Square Error (MSE) expression was also derived for the LS estimator in the case of a comb-type pilot arrangement. The MSE expression clearly explains the effects of the number of subcarriers on the performance of the LS estimator.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.497-500
• /
• 2000

In this paper, we compare the bit-error-rate performance of the fast-frequency-hopped spread-spectrum multiple-access (FFH/SSMA) and direct-sequence code-division multiple access (DS/CDMA) systems in a Rician fading channel. Each system has a same data rate, bandwidth and transmits over a Rician fading channel. The results illustrate tradoffs in performance between the FFH/SSMA and DS/CDMA systems as a function of the parameters such as average signal to noise ratio and processing gain. The performance of the FFH/SSMA system is shown to be less sensitive to the change of fading environments, while the change of processing gain and average signal to noise ratio gives considerable affect to the FFH/SSMA system compared with the DS/CDMA system. Without respect to the change of system parameters, for most of Rician fading channels (except non-fading channel), FFH/SSMA system gives better performance than DS/CDMA system in BBR < 10$\^$-3/.

• Proceedings of the KIEE Conference
• /
• 2000.07c
• /
• pp.2195-2197
• /
• 2000

In order to detect the ultrasonic that is generated by the partial discharge of the heavy electric machinery a Laser Doppler Vibrometer (LDV) is developed. A Michelson type interferometer which employed heterodyne signal process technique is built to measure the frequency and amplitude of vibration. The output signal of the fast photodetector is a frequency modulated signal centered at 40 MHz. The signal from the detector is amplified and converted to intermediate frequency centered at 1 MHz after mixing process. The voltage output that is proportional to the velocity of the moving surface(PZT) is obtained using PLL. The spectrum of the FM signal is analyzed and integration method was introduced to obtain amplitude information. This LDV can be used to measure the vibration of MEMS devices, automobiles, HDD and CDP.

• Journal of the Optical Society of Korea
• /
• v.19 no.2
• /
• pp.165-168
• /
• 2015

A simple method to determine the focal length of a lens using the Talbot image is presented. This method uses only one grating, requiring neither Moir$\acute{e}$ fringe analysis nor the angle between the gratings. The original Fourier transform was used to access the spectrum beyond the limitation set of the usual fast Fourier transform to determine the (de)magnification accurately enough to be used for the focal length. A set of Talbot images simulated numerically with the Fresnel diffraction integral was used to demonstrate the method. For focal lengths between 5550 mm and 5650 mm, the mean difference between the focal lengths determined from the Talbot images and the true values was 3.3 mm with the standard deviation of the difference being 3.8 mm. The true focal lengths can be recovered with an accuracy of 0.06%.

• ETRI Journal
• /
• v.33 no.6
• /
• pp.961-964
• /
• 2011

The cognitive ultra-wideband (UWB) network detects interfering narrowband systems and adapts its configuration accordingly. An inherently adaptive and flexible candidate for cognitive UWB transmission is the wavelet packet multicarrier modulation (WPMCM). In this letter, we use an enhanced forward consecutive mean excision thresholding algorithm to tackle the noise uncertainty in the wavelet-based sensing of WPMCM systems, and mathematical analysis is performed for primary user channel fading. As a benchmark, we compare the proposed system with a conventional fast Fourier transformation-based system, and performance investigation proves significant improvements when primary and secondary links are subjected to multipath fading and noise.

• Bulletin of the Korean Chemical Society
• /
• v.30 no.11
• /
• pp.2665-2668
• /
• 2009

A new algorithm for determining noise level is proposed for more reliability in interpreting spectral data for complex Fourier transform ion cyclotron resonance (FTICR) mass spectra of petroleum. In the new algorithm, a moving window with a fixed number of data points was adopted, instead of a fixed m/z width. In the analysis of petroleum, it was found that a moving window of 50,000 or more data points was optimal. This optimized automated peak picking performed well even with frequency-dependant noise in the mass spectrum. Additionally, this fast, automated peak picking algorithm was suitable for the analysis of a large set of samples.

• Journal of Communications and Networks
• /
• v.15 no.6
• /
• pp.601-605
• /
• 2013

The performance of M-ary turbo coded synchronous, fast frequency-hopping spread spectrum multiple-access (FHSS-MA) networks with M-ary frequency shift keying (MFSK) and noncoherent detection is analyzed under Rayleigh fading. Results indicate that M-ary turbo codes dramatically enhance the performance of FHSS-MA networks using MFSK compared to binary turbo codes.

• Preventive Nutrition and Food Science
• /
• v.15 no.4
• /
• pp.282-286
• /
• 2010

To isolate a calcium-binding peptide from chlorella protein hydrolysates, chlorella protein was extracted and hydrolyzed using Flavourzyme, a commercial protease. The degree of hydrolysis and calcium-binding capacity were determined using trinitrobenzenesulfonic acid and orthophenanthroline methods, respectively. The enzymatic hydrolysis of chlorella protein for 6 hr was sufficient for the preparation of chlorella protein hydrolysates. The hydrolysates of chlorella protein were then ultra-filtered under 5 kDa as molecular weight. The membrane-filtered solution was fractionated using ion exchange, reverse phase, normal phase chromatography, and fast protein liquid chromatography to identify a calcium-binding peptide. The purified calcium-binding peptide had a calcium binding activity of 0.166 mM and was determined to be 700.48 Da as molecular weight, and partially identified as a peptide containing Asn-Ser-Gly-Cys based on liquid chromatography/electrospray ionization tandem mass spectrum.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.1
• /
• pp.16-21
• /
• 1999

The analysis of distribution line faults is essential to the proper protections of the power system. A high impedance fault test, which was carried in Korean electric power systems, it was found that a arcing phenomenon occurred during the high level portion of conductor voltage in each cycle. In this paper, we propose a new method for detection of high impedance faults, which uses the arcing fault current difference during high voltage and low voltage portion of conductor voltage waveform. To extract this difference, we diveded one cycle fault current into equal spanned four data windows according to the magnitude of voltage waveform and applied fast fourier transform(FFT) to each data window. The frequency spectrum of current wavefrom in each portion are used as the inputs of neural network and is trained to detect high impedance faults. The proposed method shows improved accuracy when applied to staged fault data and fault-like load.