• Journal of Korea Water Resources Association
• /
• v.49 no.3
• /
• pp.207-215
• /
• 2016

In this study, it is analyzed how large scale climate variation has an effect on climate systems over Korea using correlation analysis between climate indices and Intrinsic Mode Functions (IMFs) of precipitation and temperature. For this purpose, the estimated IMFs of precipitation and temperature from the accompanying paper were used. Furthermore, cross correlation coefficients and lag time between climate indices and IMFs were calculated considering periodicities and tendencies. As results, more accurate correlation coefficients were obtained compared with those between climate indices and raw precipitation and temperature data. We found that the Korean climate is closely related with climate variations of $El-Ni{\tilde{n}}o$ in terms of periodicity and its tendency is followed with increasing sea surface temperature due to climate change.

• Korean Journal of Optics and Photonics
• /
• v.5 no.2
• /
• pp.326-332
• /
• 1994

Performance of soliton based long-haul optical communication system with in-line optical amplifier repeaters are evaluated numerically. To reduce the optical amplifier noise, the amplifier gain is increased and the post amplifier loss is included. By theoretical calculation with 1480 nm co propagating pump, 7 dB amplifier gain and operating the amplifier 1 dB in compression, the spontaneous emission factor can be reduced from 2.37 to 1.45 by increasing the pump power from 3.71 to 11.53 mW and increasing the post amplifier loss from 0 to 10 dB. Then, power penalty can be reduced from 4.09 to 1.20 dB for 8,000'km transmission and the maximum transmission distance is 14,890 km. ,890 km.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.5
• /
• pp.1990-1995
• /
• 2012

This paper presents design methodology to determine the design parameters that affect the manufacture of aluminum forging piston using the FE simulation and the Taguchi method. Maximum forging load is used as the objective function, and preform, material temperature and draft angle are selected as the design parameters. Their combinations are implemented by orthogonal array, and forging load is evaluated through the simulation. From the analytic results of design parameters to minimize the load using signal to noise ratio, their optimal combinations are proposed. The proposed design methodology will be able to help in selecting proper preform among preforms and to be used in determining the optimal combination of the parameters in metal forming process.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.1B
• /
• pp.89-98
• /
• 2000

This paper presents an active noise cancelling model using wavelet transform and subband filter banks based on adaptive filter. The analysis filter banks decompose input and error signals into QMF filter banks of lowpass and highpass bands. Each filter bank uses wavelet filter with dyadic tree structure. The decomposed input and error signals are iterated by adaptive filter coefficients of each subband using filtered-X LMS algorithm. The synthesis filter banks make output signal of wideband with perfect reconstruction to prepare adaptive filter output signals of each subband. The analysis and synthesis niter hants use conjugate quadrature filters for Pefect reconstruction. Also, The delayed LMS algorithm model for on-line identification of error path transfer characteristics is used gain and acoustic time delay factors. The proposed adaptive active noise cancelling modelis suggested by system retaining the computational and convergence speed advantage using wavelet subband filter banks.

• Journal of the Korean Society of Radiology
• /
• v.17 no.1
• /
• pp.141-148
• /
• 2023

The algebraic reconstruction technique is one of the reconstruction methods in CT and shows good image quality against noise-dominant conditions. The number of iteration is one of the key factors determining the execution time for the algebraic reconstruction technique. However, there are some rules for determining the number of iterations that result in more than a few hundred iterations. Thus, the rules are difficult to apply in practice. In this study, we proposed a method to determine the number of iterations for practical applications. The reconstructed image quality shows slow convergence as the number of iterations increases. Image quality 𝜖 < 0.001 was used to determine the optimal number of iteration. The Shepp-Logan head phantom was used to obtain noise-free projection and projections with noise for 360, 720, and 1440 views were obtained using Geant4 Monte Carlo simulation that has the same geometry dimension as a clinic CT system. Images reconstructed by around 10 iterations within the stop condition showed good quality. The method for determining the iteration number is an efficient way of replacing the best image-quality-based method, which brings over a few hundred iterations.

• Journal of the Korean Society of Radiology
• /
• v.15 no.6
• /
• pp.927-933
• /
• 2021

The purpose of this study was to compare and analyze the differences in scan time, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) in the third lumbar vertebral region including the back fat, spinal cord, and cerebrospinal fluid using the mDixon, T2 TSE, and T2 spectral pre-saturation with inversion-recovery (SPIR) techniques. With the factors affecting the SNR fixed, the lumbar sagittal plane images of 30 adults were compared on mDixon, T2 TSE, and T2 SPIR imaging tests. The test times for mDixon, T2 TSE, and T2 SPIR were 115 seconds, 60 seconds, and 60 seconds, respectively. The mDixon T2 images showed higher SNR than the T2 TSE images at the third lumbar vertebral region (p<0.05), lower SNR in the back fat and cerebrospinal fluid (p<0.05) areas, and comparable SNR in the spinal cord (p>0.05). The CNR between the third lumbar vertebral area and back fat was higher in the mDixon T2 images, and the CNR of the cerebrospinal fluid and spinal cord images was higher in the T2 TSE images (p<0.05). The mDixon T2 FS images CNR was lower for the 3rd lumbar vertebral body region and back fat than the T2 SPIR images, and higher for the spinal cord and cerebrospinal fluid images (p<0.05). The CNR between the third lumbar body and back fat areas was higher in the mDixon T2 FS images (p<0.05), and there was no difference in the CNR in the images of the cerebrospinal fluid and the spinal cord (p>0.05). It is difficult to determine whether the mDixon technique is superior to the conventional T2 TSE and T2 SPIR techniques in terms of test time, SNR, and CNR. This study was confined to patients with simple lower back pain and was limited by controlled experimental conditions. Studies using clinically applied protocols are warranted in the future.

• Investigative Magnetic Resonance Imaging
• /
• v.13 no.2
• /
• pp.161-170
• /
• 2009

Purpose : A parallel imaging method provides us to improve temporal resolution to obtain three-dimensional (3D) MR images. The objective of this study was to optimize three 3D MRI techniques by adjusting 2D SESNE factors of the parallel imaging method in phantom and human brain. Materials and Methods : With a 3 Tesla MRI system and an 8-channel phase-array sensitivity-encoding (SENSE) coil, three 3D MRI techniques of 3D T1-weighted imaging (3D T1WI), 3D T2-weighted imaging (3D T2WI) and 3D fluid attenuated inversion recovery (3D FLAIR) imaging were optimized with adjusting SESNE factors in a water phantom and three human brains. The 2D SENSE factor was applied on the phase-encoding and the slice-encoding directions. Signal-to-noise ratio(SNR), percent signal reduction rate(%R), and contrast-to-noise ratio(CNR) were calculated by using signal intensities obtained in specific regions-of-interest (ROI). Results : In the phantom study, SENSE factor = 3 was provided in 0.2% reduction of signals against without using SENSE with imaging within 5 minutes for 3D T1WI. SENSE factor = 2 was provided in 0.98% signal reduction against without using SENSE with imaging within 5 minutes for 3D T2WI. SENSE factor = 4 was provided in 0.2% signal reduction against without using SENSE with imaging around 6 minutes for 3D FLAIR. In the human brain study, SNR and CNR were higher with SENSE factors = 3 than 4 for all three imaging techniques. Conclusion : This study was performed to optimize 2D SENSE factors in the three 3D MRI techniques that can be scanned in clinical time limitations with minimizing SNR reductions. Without compromising SNR and CNR, the optimum 2D SENSE factors were 3 and 4, yielding the scan time of about 5 to 6 minutes. Further studies are necessary to optimize 3D MRI techniques in other areas in human body.

• Geophysics and Geophysical Exploration
• /
• v.20 no.3
• /
• pp.176-184
• /
• 2017

To interpret the seismic image precisely, random noises should be suppressed and the continuity of the image should be enhanced by using the appropriate smoothing techniques. Structure-Oriented Filter-Edge Preserving (SOF-EP) technique is one of the methods, that have been actively researched and used until now, to efficiently smooth seismic data while preserving the continuity of signal. This technique is based on the principle that diffusion occurs from large amplitude to small one. In a continuous structure such as a horizontal layer, diffusion or smoothing is operated along the layer, thereby increasing the continuity of layers and eliminating random noise. In addition, diffusion or smoothing across boundaries at discontinuous structures such as faults can be avoided by employing the continuity decision factor. Accordingly, the precision of the smoothing technique can be improved. However, in the case of the structure-oriented semblance technique, which has been used to calculate the continuity factor, it takes lots of time depending on the size of the filter and data. In this study, we first implemented the SOF-EP method and confirmed its effectiveness by applying it step by step to the field data. Next, we proposed and applied the corner response method which can efficiently calculate the continuity decision factor instead of structure-oriented semblance. As a result, we could confirm that the computation time can be reduced by about 6,000 times or more by applying the corner response method.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.414-414
• /
• 2011

우리나라 뿐만 아니라 전 세계적으로 기후변화로 인한 집중호우, 폭설 등이 빈번하게 일어나고 있으며 수공구조물 설계에 필요한 확률강우량도 증가하고 있다. 확률강우량을 산정하는 빈도해석의 경우 지점빈도해석의 문제점을 보완한 지역빈도해석에 대한 연구가 꾸준히 진행되고 있다. 지역빈도해석을 적용하기 위해서는 수문학적 동질성을 가지는 지역 구분이 무엇보다 중요하다. 군집 분석은 개체들이 지니고 있는 다양한 속성의 유사성을 동질적인 집단으로 군집화하는 방법을 말한다. 군집분석의 기본원리는 분석하고자 하는 여러 특성등을 유사성(similaruty) 거리(distance)로 환산하고 거리가 상대적으로 가까운 개체들을 동질적으로 군집화하는 것이다. 군집분석을 적용하기 위해서는 기상학적 인자와 지형학적 인자를 이용하여 군집분석을 실시한다. 군집분석을 실시할 때 가장 중요한 것은 입력변수의 선택으로 입력 변수의 적절한 선택이 결과값에 큰 영향을 준다. 상호정보량(Mutual Information, MI) 기법은 두 무작위 변수간의 관련성을 측정하는 방법이며 (Cover and Tomas, 2006), 두 변수간의 독립성 구조에 관한 가정이 없고 데이터 변형이나 잡음(noise)에 대한 영향이 적어 다른 기법보다 신뢰도가 높다고 알려져 있다(Peng et al., 2005). 본 연구에서는 상호정보량 기법을 이용하여 군집된 지점들의 종속성과 독립성의 관계를 정량적으로 산정하여 비교하고자 한다.

• The Journal of Korean Institute of Information Technology
• /
• v.15 no.12
• /
• pp.157-164
• /
• 2017

Mobile ECG signal measurement is a technique to measure small signals of several mV, and many studies have been conducted to remove noise including wandering scheme. Removal of the equipotential line noise caused by shaking or movement of the electrode cable is one of the core research contents for the electrocardiogram measurement. In this study, we proposed a modified step-size of combined NLMS(normalized least squares) and DLMS(delayed least squares) adaptive filter to eliminate baseline noise from ECG signals. The proposed method mainly adjusts initial filter step-size to reduce distortion of original ECG signals characteristic after eliminating baseline noise. The modified filter step-size is scaled by filter order size and distortion minimization factor. This method is suitable for portable ECG device with a small processor and less power consumption. This technique also decreases computation time which is essential for real-time filtering. The proposed filter also increase the signal to noise ratio (SNR) compared to conventional NLMS filter.