• Journal of the Society of Naval Architects of Korea
• /
• v.51 no.5
• /
• pp.409-418
• /
• 2014

Simulation of flow past a complex marine structure requires a fine resolution in the vicinity of the structure, whereas a coarse resolution is enough far away from it. Therefore, a lot of grid cells may be wasted, when a simple Cartesian grid system is used for an Immersed Boundary Method (IBM). To alleviate this problems while maintaining the Cartesian frame work, we adopted an Adaptive Mesh Refinement (AMR) scheme where the grid system dynamically and locally refines as needed. In this study, We implemented a moving IBM and an AMR technique in our basic 3D incompressible Navier-Stokes solver. A Volume Of Fluid (VOF) method was used to effectively treat the free surface, and a recently developed Lagrangian Dynamic Subgrid-scale Model (LDSM) was incorporated in the code for accurate turbulence modeling. To capture vortex induced vibration accurately, the equation for the structure movement and the governing equations for fluid flow were solved at the same time implicitly. Also, We have developed an interface by using AutoLISP, which can properly distribute marker particles for IBM, compute the geometrical information of the object, and transfer it to the solver for the main simulation. To verify our numerical methodology, our results were compared with other authors' numerical and experimental results for the benchmark problems, revealing excellent agreement. Using the verified code, we investigated the following cases. (1) simulating flow around a floating sphere. (2) simulating flow past a marine structure.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.5
• /
• pp.425-430
• /
• 2015

This paper is a study to improve the classification efficiency of rotating objects by using deep neural networks to which a deep learning algorithm was applied. For the classification experiment of rotating objects, COIL-20 is used as data and total 3 types of classifiers are compared and analyzed. 3 types of classifiers used in the study include PCA classifier to derive a feature value while reducing the dimension of data by using Principal Component Analysis and classify by using euclidean distance, MLP classifier of the way of reducing the error energy by using error back-propagation algorithm and finally, deep learning applied DBN classifier of the way of increasing the probability of observing learning data through pre-training and reducing the error energy through fine-tuning. In order to identify the structure-specific error rate of the deep neural networks, the experiment is carried out while changing the number of hidden layers and number of hidden neurons. The classifier using DBN showed the lowest error rate. Its structure of deep neural networks with 2 hidden layers showed a high recognition rate by moving parameters to a location helpful for recognition.

• Journal of Korean Society for Geospatial Information Science
• /
• v.14 no.3 s.37
• /
• pp.71-77
• /
• 2006

MSAS satellite is the geostationary satellite for realizing WADGPS that can get the position of moving object in a wide area receiving the correction signal created from a ground using satellite. In this study, we analyzed two different data. One is using the correction signal transmitted from MTSAT-2 satellite of MSAS and the other is receiving the data of DGPS using BEACON receiver. As we compared both data, we could get the conclusion that the position accuracy of both data is also can get up to the standard or the conventional real-time code DGPS. As a result, we can expect that if we use MTSAT-2 satellite and BEACON receiver together, we can apply them LBS part that require real-time data or the obtaining geospatial information that does not require high accuracy much regardless of topography.

• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.5
• /
• pp.480-488
• /
• 2006

Industrial and economical importance of CP(Cellular Phone) is growing rapidly. Combined with IT technology, CP is one of the most attractive technologies of today. However, unless we find a new breakthrough in the technology, its growth may slow down soon. RT(Robot Technology) is considered one of the most promising next generation technologies. Unlike the industrial robot of the past, today's robots require advanced features, such as soft computing, human-friendly interface, interaction technique, speech recognition object recognition, among many others. In this paper, we present a new technological concept named RCP (Robotic Cellular Phone) which integrates RT and CP in the vision of opening a combined advancement of CP, IT, and RT, RCP consists of 3 sub-modules. They are $RCP^{Mobility}$(RCP Mobility System), $RCP^{Interaction}$, and $RCP^{Integration}$. The main focus of this paper is on $RCP^{Mobility}$ which combines an autonomous navigation system of the RT mobility with CP. Through $RCP^{Mobility}$, we are able to provide CP with robotic functions such as auto-charging and real-world robotic entertainment. Ultimately, CP may become a robotic pet to the human beings. $RCP^{Mobility}$ consists of various controllers. Two of the main controllers are trajectory controller and self-localization controller. While the former is responsible for the wheel-based navigation of RCP, the latter provides localization information of the moving RCP With the coordinates acquired from RFID-based self-localization controller, trajectory controller refines RCP's movement to achieve better navigation. In this paper, a prototype of $RCP^{Mobility}$ is presented. We describe overall structure of the system and provide experimental results on the RCP navigation.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.10 no.3
• /
• pp.91-95
• /
• 2010

Using artificial neural network to use a search preference based on the user's information, the ranking of search results that will enable flexible searches can be improved. After trained in several different queries by other users in the past, the actual search results in order to better reflect the use of artificial neural networks to neural network learning. In order to change the weights constantly moving backward in the network to change weights of backpropagation algorithm. In this study, however, the initial training, performance data, look for increasing the number of lessons that can be overfitted. In this paper, we have optimized a lot of objects that have a strong advantage to apply genetic algorithms to the relevant page of the search rankings flexible as an object to the URL list on a random selection method is proposed for the study.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.6
• /
• pp.153-162
• /
• 2009

This paper suggests the novel light-weight signal processing algorithm for wireless sensor network applications which needs low computing complexity and power consumption. Exponential average method (EA) is utilized by real time, to process the magnetometer signal which is analyzed to understand the own physical characteristic in time domain. EA provides the robustness about noise, magnetic drift by temperature and interference, furthermore, causes low memory consumption and computing complexity for embedded processor. Hence, optimal parameter of proposal algorithm is extracted by statistical analysis. Using general and precision magnetometer, detection probability over 90% is obtained which restricted by 5% false alarm rate in simulation and using own developed magnetometer H/W, detection probability over 60~70% is obtained under 1~5% false alarm rate in simulation and experiment.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.5
• /
• pp.25-34
• /
• 2008

By comparing two images, which are captured with the same scene at different time, we can detect a set of outliers, such as occluding objects due to moving vehicles. To reduce the influence from the different intensity properties of the images, an intensity compensation scheme, which is based on the polynomial regression model, is employed. For an accurate detection of outliers alleviating the influence from a set of outliers, a simple technique that reruns the regression is employed. In this paper, an algorithm that iteratively reruns the regression is theoretically analyzed by observing the convergence property of the estimates of the noise variance. Using a correction constant for the estimate of the noise variance is proposed. The correction enables the detection algorithm robust to the choice of thresholds for selecting outliers. Numerical analysis using both synthetic and Teal images are also shown in this paper to show the robust performance of the detection algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.1A
• /
• pp.80-88
• /
• 2002

Visual media(image, graphic, and video) processing poses challenge from several perpectives, specifically from the point of view of real-time implementation and scalability. There have been several approaches to obtain speedups to meet the computing demands in multimedia processing ranging from media processors to special purpose implementations. A variety of parallel processing strategies are adopted in these implementations in order to achieve the required speedups. We have investigated a parallel processing system for improving the processing speed o f visual media related applications. The parallel processing system we proposed is similar to a pipelined memory stystem(MAMS). The multi-access memory system is made up of m memory modules and a memory controller to perform parallel memory access with a variety of combinations of 1${\times}$pq, pq${\times}$1, and p${\times}$q subarray, which improves both cost and complexity of control. Facial recognition, Phong shading, and automatic segmentation of moving object in image sequences are some that have been applied to the parallel processing system and resulted in faithful processing speed. This paper describes the parallel processing systems for the speedup and its utilization to three time-consuming applications.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.3A
• /
• pp.274-281
• /
• 2006

Sensor network applications need synchronized time extremely such as object tracking, consistent state updates, duplicate detection, and temporal order delivery. This paper describes reliable time synchronization protocol (RTSP) for wireless sensor networks. In the proposed method, synchronization error is decreased by creating hierarchical tree with lower depth and reliability is improved by maintaining and updating information of candidate parent nodes. The RTSP reduces recovery time and communication overheads comparing to TPSN when there are topology changes owing to moving of nodes, running out of energy and physical crashes. Simulation results show that RTSP has about 20% better performance than TPSN in synchronization accuracy. And the number of message in the RTSP is $20%{\sim}60%$ lower than that in the TPSN when nodes are failed in the network. In case of different transmission range of nodes, the communication overhead in the RTSP is reduced up to 40% than that in the TPSN at the maximum.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.2
• /
• pp.83-88
• /
• 2006

Many applications in the area of location-based services and personal navigation require nowadays the location determination of a user not only in outdoor environment but also indoor. To locate a person or object in a building, systems that use either infrared, ultrasonic or radio signals, and visible light for optical tracking have been developed. The use of WiFi for location determination has the advantage that no transmitters or receivers have to be installed in the building like in the case of infrared and ultrasonic based location systems. WiFi positioning technology adopts IEEE802.11x standard, by observing the radio signals from access points installed inside a building. These access points can be found nowadays in our daily environment, e.g. in many office buildings, public spaces and in urban areas. The principle of operation of location determination using WiFi signals is based on the measurement of the signal strengths to the surrounding available access points at a mobile terminal (e.g. PDA, notebook PC). An estimate of the location of the terminal is then obtained on the basis of these measurements and a signal propagation model inside the building. The signal propagation model can be obtained using simulations or with prior calibration measurements at known locations in an offline phase. The most common location determination approach is based on signal propagation patterns, namely WiFi fingerprinting. In this paper the underlying technology is briefly reviewed followed by an investigation of two WiFi positioning systems. Testing of the system is performed in two localization test beds, one at the Vienna University of Technology and the second at the Hong Kong Polytechnic University. First test showed that the trajectory of a moving user could be obtained with a standard deviation of about ${\pm}$ 3 m.