• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.03a
• /
• pp.188-194
• /
• 2009

Consolidation settlement, a crucial parameter in geotechnical design of soft ground, has not been computed in a unique way due to different computation methods in practice. To improve computational error in calculating consolidation settlement, a number of researches has been attempted. Conventional 1-dimensional consolidation theory assumes the center of the clay layer as the representative point to obtain effective stress in calculation, which could resort to erroneous results. To calculate exact solutions considering initial distribution of effective stress, diving a stratum into multi-layers could resort to wasting time and effort. In the study, a novel methodology for calculating consolidation settlement via Guassian quadrature is developed. The method generally is capable of computing settlements in any case of the stress conditions encountered in fields.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.27 no.2
• /
• pp.9-15
• /
• 1990

A soft-decision decoding algorithm for linear binary block codes is proposed, for minimizing the block error probability. To compare the proposed algorithm with already established decoding methods, computer simulations are performed for the (7,4)Hamming code and the (23,12) Golay code. The average number of hard-decision decoding is always less then 2, and approaches to 1 when the signal to noise ratio is sufficiently large. These results show that the proposed algorithm reduces the decoding complexity.

• Annual Conference on Human and Language Technology
• /
• 2006.10e
• /
• pp.25-31
• /
• 2006

본 논문에서는 띄어쓰기 오류와 철자 오류를 동시에 교정 가능한 전처리기를 제안한다. 제시된 알고리즘은 기존의 전처리기 알고리즘이 각 오류를 따로 해결하는 데에서 오는 한계를 극복하고, 기존의 noisy-channel model을 확장하여 대화체의 띄어쓰기 오류와 철자오류를 동시에 효과적으로 교정할 수 있다. N-gram과 자소변환확률 등의 통계적 방법과 어절변환패턴 사전을 이용하여 최대한 사전을 적게 이용하면서도 효과적으로 교정 후보들을 생성할 수 있다. 실험을 통해 현재 단계에서는 만족할 만한 성능을 얻지는 못하였지만 오류 분석을 통하여 이와 같은 방법론이 실제로 효용성이 있음을 알 수 있었고 앞으로 더 많은 개선을 통해 일상적인 대화체 문장에 대해서 효과적인 전처리기로서 기능할 수 있을 것으로 기대 된다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.5A
• /
• pp.487-493
• /
• 2008

In this paper, we consider coded layered space-time architecture in MIMO-OFDM channels. Vertical Bell Lab Layered Space-Time(V-BLAST) scheme employing ordered successive interference cancellation(OSIC) algorithm provides very high spectral efficiency with low computational complexity. However, the error propagation is a major drawback constraining the overall performance of the V-BLAST system significantly. Based on this problem, we derive an improved detector using soft bit log-likelihood ratio(LLR) value. Simulation results show that the proposed detector outperforms the conventional V-BLAST scheme under spatially uncorrelated as well as correlated fading channels.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.15 no.1
• /
• pp.21-28
• /
• 2004

In this paper we apply block turbo coding at the transmitter and iterative decoding algorithm at the receiver for different operating modes, based on the 5 ㎓ IEEE 802.1 la WLAN system, instead of convolutional coding and soft decision viterbi algorithm to improve forward error correcting performance. Experimental results showed that each coding scheme outperforms coding gains of up to 3.5 ㏈ at the BER of 10$\^$-3/.

• Journal of Internet Computing and Services
• /
• v.11 no.1
• /
• pp.21-26
• /
• 2010

In this paper, film uniformity is improved by the bolt-mapping of convolutional image processing between bolts and a film using soft X-ray based in-line LCD TV film thickness controls. The automatic film thickness analysis of 3 inter-bolt's convolution function is proposed for the reduction of offset error from the X-ray changing of line scan in film profiles.

• Journal of Semiconductor Engineering
• /
• v.1 no.3
• /
• pp.81-87
• /
• 2020

An instrumentation amplifier (IA) and an analog-to-digital converter (ADC) are essential circuit blocks for accurate and robust sensor readout systems. This paper introduces recent advances in radiation-hardening by design (RHBD) techniques applied for the sensor readout integrated circuits (IC), e.g., the three-op-amp IA and the successive-approximation register (SAR) ADC, operating against total ionizing dose (TID) and singe event effect (SEE) in harsh radiation environments. The radiation-hardened IA utilized TID monitoring and adaptive reference control to compensate for transistor parameter variations due to radiation effects. The radiation-hardened SAR ADC adopts delay-based double-feedback flip-flops to prevent soft errors which flips the data bits. Radiation-hardened IA and ADC were verified through compact model simulation, and fabricated CMOS chips were measured in radiation facilities to confirm their radiation tolerance.

• The Journal of Korea Robotics Society
• /
• v.19 no.1
• /
• pp.92-97
• /
• 2024

This paper provides a study of distributional perspective on reinforcement learning for application in mobile robot navigation. Mapless navigation algorithms based on deep reinforcement learning are proven to promising performance and high applicability. The trial-and-error simulations in virtual environments are encouraged to implement autonomous navigation due to expensive real-life interactions. Nevertheless, applying the deep reinforcement learning model in real tasks is challenging due to dissimilar data collection between virtual simulation and the physical world, leading to high-risk manners and high collision rate. In this paper, we present distributional reinforcement learning architecture for mapless navigation of mobile robot that adapt the uncertainty of environmental change. The experimental results indicate the superior performance of distributional soft actor critic compared to conventional methods.

• Analytical Science and Technology
• /
• v.18 no.1
• /
• pp.5-12
• /
• 2005

It has been established that soft error of high precision electronic circuits can be induced by alpha particles emitted from the naturally occurring radioactive impurities such as U, and Th. As the electronic circuits have recently become lower dimension and higher density, these alpha-particle emitting radioactive impurities have to be strictly controlled. The aim of this study is to develop of NAA (Neutron Activation Analysis) and gamma-spectrometry to improve the analytical sensitivity and precision of U and Th. A new NAA method has been established using the HTS (Hydrulic transfer system) irradiation facility which has been used to produce radioisotopes for industries and medicines instead of the PTS (pneumatic transfer system) irradiation facility which has been used in general NAA. When the ultratrace impurities have to be analyzed by NAA, background gamma-ray spectra induced from $^{222}Rn$ and its progenies in air is serious problem. This unstable background has been eliminated or stabilized by the use of a nitrogen purging system. Ultra trace amounts of U (0.1 ng/g) and Th (0.01 ng/g) in high purity silica used for EMC could be analyzed by the use of HTS-NAA and low background gamma-spectrometry.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.44 no.1
• /
• pp.138-145
• /
• 2007

Wireless transmission system using the multipath channel is affected ISI due to the delay spread. So we use a decision feedback equalizer which consist of decision part and feedback filter for remove the ISI effectively. In this paper, we propose a improved adaptive decision feedback equalizer to mitigate ISI effectively. The proposed adaptive decision feedback equalizer is construct by using soft decision device and hierarchical feedback filter based on MMSE sub-optimal equalizer using the LMS algorithm. Soft decision device mitigate the error propagation in feedback filter by incorrectly detected decision symbol and feedback filter which is divided two step independently mitigate the ISI by using a adaptive algorithm. As a result this structure shows better performance than conventional decision feedback equalizer by mitigating the error propagation in filter cause incorrectly detecting symbol. and we get the MSE more rapidly by using larger step-size due to reduce the number of feedback filter tap. In computer simulation, we compare the bit error rate performance of proposed decision feedback equalizer with conventional one on the S-V channel model for UWB system.