• The Journal of the Acoustical Society of Korea
• /
• v.24 no.8
• /
• pp.454-461
• /
• 2005

Time spread due to the multipath in underwater acoustic channel, induces ISI (Inter-Symbol Interference) which degrades the performance of the underwater acoustic communication system. The passive phase conjugation (PPC) which improves the signal to multipath interference ratio (SMR) and therefore reduces the frequency selectivity. is a diversity communication technique giving a less ISI under multipath fading channel. Its frequency selectivity depends on the number of receiver array sensors and time varying source to receiver range. In this study, frequency selectivity of the PPC and its effects on bit error of underwater acoustic communication is analyzed by numerical simulation.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.49 no.4
• /
• pp.61-68
• /
• 2012

In this paper, a closed-loop sensorless stroke control system for a linear compressor has been designed. In order to estimate the piston position accurately, motor parameters are identified as a function of the piston position and the motor current. These parameters are stored in ROM table and used later for the accurate estimation of piston position. The identified motor parameters are approximated to the several surface functions in order to decrease memory size. They can also be divided into 2 or 4 subsections to decrease identification errors. The effect of the order of surface functions and division of subsections on identification errors and computation time is analyzed.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.13 no.1
• /
• pp.55-62
• /
• 2008

This paper proposes a feedback linearization control scheme of AC/DC PWM converters with LCL input filters using no damping resisters. This feedback linearization scheme can eliminate the non-linearity of the system. So, the controller of the system can be designed by using linear control theory, which gives a good transient response. The cascade structure of the controller makes the converter current be controlled within a certain limit. To reduce the number of sensors, the source voltage and current is estimated. The validity of the proposed control algorithm is verified by simulation and experimental results.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.4
• /
• pp.497-502
• /
• 2010

The accurate estimation of position and direction of the mobile robot is essential for preparing precise movement and works in the inner complex environment. In this paper, we propose a robust estimation method of location and direction using the velocity of mobile robot in the inner GPS environment. The estimation using the inner GPS with ultrasonic sensors have to consider with various acoustic noise and sensor errors. We design a robust estimation method using a membership function based on uncertainty of the obtained information and robot velocity. The simulation results of the proposed method show effectiveness in the contaminated environment with position errors.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.229-229
• /
• 2023

강화학습은 에이전트(agent)가 주어진 환경(environment)과의 상호작용을 통해서 상태(state)를 변화시켜가며 최대의 보상(reward)을 얻을 수 있도록 최적의 행동(action)을 학습하는 기계학습법을 의미한다. 최근 알파고와 같은 게임뿐만 아니라 자율주행 자동차, 로봇 제어 등 다양한 분야에서 널리 사용되고 있다. 상수도관망 분야의 경우에도 펌프 운영, 밸브 운영, 센서 최적 위치 선정 등 여러 문제에 적용되었으나, 설계에 강화학습을 적용한 연구는 없었다. 설계의 경우, 관망의 크기가 커짐에 따라 알고리즘의 탐색 공간의 크기가 증가하여 기존의 최적화 알고리즘을 이용하는 것에는 한계가 존재한다. 따라서 본 연구는 강화학습을 이용하여 상수도관망의 구성요소와 환경요인 간의 복잡한 상호작용을 고려하는 설계 방법론을 제안한다. 모델의 에이전트를 딥 강화학습(Deep Reinforcement Learning)으로 구성하여, 상태 및 행동 공간이 커 발생하는 고차원성 문제를 해결하였다. 또한, 해당 모델의 상태 및 보상으로 절점에서의 압력 및 수요량과 설계비용을 고려하여 적절한 수량과 수압의 용수 공급이 가능한 경제적인 관망을 설계하도록 하였다. 모델의 행동은 실제로 공학자가 설계하듯이 절점마다 하나씩 차례대로 다른 절점과의 연결 여부를 결정하는 것으로, 이를 통해 관망의 레이아웃(layout)과 관경을 결정한다. 본 연구에서 제안한 방법론을 규모가 큰 그리드 네트워크에 적용하여 모델을 검증하였으며, 고려해야 할 변수의 개수가 많음에도 불구하고 목적에 부합하는 관망을 설계할 수 있었다. 모델 학습과정 동안 에피소드의 평균 길이와 보상의 크기 등의 변화를 비교하여, 제안한 모델의 학습 능력을 평가 및 보완하였다. 향후 강화학습 모델을 통해 신뢰성(reliability) 또는 탄력성(resilience)과 같은 시스템의 성능까지 고려한 설계가 가능할 것으로 기대한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.259-259
• /
• 2022

하천에서 측정된 표면유속을 이용하여 유량을 산정하는 방법은 전통적으로는 여러 개의 측선에서 표면유속을 측정하고 해당 단면적을 적용하여 전체의 유량을 산정한다. 이러한 방법은 조사자가 직접 표면유속계를 이용하여 이동하면서 측정해야 하며, 측정에 상당한 시간이 소요된다. 또한 사람이 직접 교량 위, 케이블 등을 통해 측정해야 하므로 위험에 노출될 수 있다. 따라서 실시간 표면유속을 측정하여 무인으로 유량을 산정할 수 있는 방안이 필요하다. 본 연구에서는 최대유속이 발생하는 하나의 지점에서 측정된 표면유속을 이용하여 전체의 유량을 산정할 수 있도록 지표유속법(index velocity method)과 유속분포법(velocity distribution method)을 적용하여 비교하고자 한다. 지표유속법은 최대유속과 평균유속과의 관계식을 작성하여 유량을 산정하는 방식이며, 유속분포법은 Chiu(1987, 1988)의 엔트로피 개념의 유속분포법(개수로 단면에서의 2차원적 유속분포)을 적용하고자 한다. 연구를 위해 사용되는 자료는 8개 지점에서 전자파표면유속계로 측정된 유속 및 유량자료를 이용하였다. 본 연구결과를 활용하여 표면유속을 이용한 자동유량측정시스템을 도입한다면 자동유량측정시설의 설치 및 운영에 있어 기존 수중에 센서를 설치하여 운영하는 것보다 효율적일 것으로 판단된다. 또한 단면형상 및 흐름이 복잡하지 않는 하천에서 활용도가 높을 것으로 판단된다.

• Journal of the Korea Computer Graphics Society
• /
• v.25 no.3
• /
• pp.85-92
• /
• 2019

A 3D virtual character playing a role in a digital story-telling has a unique style in its appearance and motion. Because the style reflects the unique personality of the character, it is very important to preserve the style and keep its consistency. However, when the character's motion is directly controlled by a user's motion who is wearing motion sensors, the unique style can be discarded. We present a novel character motion control method that uses only a small amount of animation data created only for the character to preserve the style of the character motion. Instead of machine learning approaches requiring a large amount of training data, we suggest a search-based method, which directly searches the most similar character pose from the animation data to the current user's pose. To show the usability of our method, we conducted our experiments with a character model and its animation data created by an expert designer for a virtual reality game. To prove that our method preserves well the original motion style of the character, we compared our result with the result obtained by using general human motion capture data. In addition, to show the scalability of our method, we presented experimental results with different numbers of motion sensors.

• Korean Journal of Remote Sensing
• /
• v.29 no.2
• /
• pp.183-195
• /
• 2013

Real-time high-precision image georeferencing is important for the realization of image based precise navigation or sophisticated augmented reality. In general, high-precision image georeferencing can be achieved using the conventional simultaneous bundle adjustment algorithm, which can be performed only as post-processing due to its processing time. The recently proposed sequential bundle adjustment algorithm can rapidly produce the results of the similar accuracy and thus opens a possibility of real-time processing. However, since the processing time still increases linearly according to the number of images, if the number of images are too large, its real-time processing is not guaranteed. Based on this algorithm, we propose a modified fast algorithm, the processing time of which is maintained within a limit regardless of the number of images. Since the proposed algorithm considers only the existing images of high correlation with the newly acquired image, it can not only maintain the processing time but also produce accurate results. We applied the proposed algorithm to the images acquired with 1Hz. It is found that the processing time is about 0.02 seconds at the acquisition time of each image in average and the accuracy is about ${\pm}5$ cm on the ground point coordinates in comparison with the results of the conventional simultaneous bundle adjustment algorithm. If this algorithm is converged with a fast image matching algorithm of high reliability, it enables high precision real-time georeferencing of the moving images acquired from a smartphone or UAV by complementing the performance of position and attitude sensors mounted together.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.5
• /
• pp.143-151
• /
• 2016

The aim of this study is to investigate the optimal algorithm to extract medical radiation induced pixel signal from complementary metal-oxide semiconductor (CMOS) sensors of smartphones camera. The pixel intensity and pixel number of smartphone camera were measured as the X-ray dose was increased. The front camera of the smartphone camera has low noise property and excellent dose response as compared to the back camera of the smartphone. The indirect method which uses scintillation crystal in front of the smartphone camera, couldn't improve the X-ray detection efficiency as compared to the direct method which does not use any scintillator in front of the smartphone camera. When we used the algorithm which employing threshold level on the pixel intensity and pixel number, the dose linearity was more higher for the pixel intensity rather for the pixel number. The use of pixel intensity of Y color component which represents the grey scale, would be efficient in terms of the radiation detection efficiency and reducing the complexity of the image processing. We expect that the radiation dose monitoring can be managed effectively and systematically by using the proposed radiation detection algorithm, thus eventually will contribute to the public healthcare.

• Journal of the Korean Magnetics Society
• /
• v.25 no.4
• /
• pp.117-122
• /
• 2015

The holes with a diameter of $35{\mu}m$ inside the GMR-SV (giant magnetoresistance-spin valve) film were patterned by using the photolithography process and ECR (electron cyclotron resonance) Ar-ion milling. From the magnetoresistance curves of the GMR-SV film with holes measuring by 4-electrode method, the MR (magnetoresistance ratio) and MS (magnetic sensitivity) are almost same as the values of initial states. On other side hand, the $H_{ex}$ (exchange bias coupling field) and $H_c$ (coercivity) dominantly increased from 120 Oe and 10 Oe to 190 Oe and 41 Oe as increment of the number of holes inside GMR-SV film respectively. These results were shown to be attributed to major effect of EMD (easy magnetic domian) having a region positioned between two holes perpendicular to the sensing current. On the basis of this study, the fabrication of GMR-SV applying to the hole formation improved the magnetoresistance properties having the thermal stability and durability of bio-device.