• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.307-325
• /
• 2023

Due to the prolonged COVID-19 pandemic, the frequency of people who are tired of living indoors visiting nearby mountains and national parks to relieve depression and lethargy has exploded. There is a place where thousands of people who came out of nature stop walking and breathe and rest, that is the mineral spring. Even in mountains or national parks, there are about 600 mineral springs that can be found occasionally in neighboring parks or trails in the metropolitan area. However, due to irregular and manual water quality tests, people drink mineral water without knowing the test results in real time. Therefore, in this study, we intend to develop a model that can predict the quality of the spring water in real time by exploring the factors affecting the quality of the spring water and collecting data scattered in various places. After limiting the regions to Seoul and Gyeonggi-do due to the limitations of data collection, we obtained data on water quality tests from 2015 to 2020 for about 300 mineral springs in 18 cities where data management is well performed. A total of 10 factors were finally selected after two rounds of review among various factors that are considered to affect the suitability of the mineral spring water quality. Using AutoML, an automated machine learning technology that has recently been attracting attention, we derived the top 5 models based on prediction performance among about 20 machine learning methods. Among them, the catboost model has the highest performance with a prediction classification accuracy of 75.26%. In addition, as a result of examining the absolute influence of the variables used in the analysis through the SHAP method on the prediction, the most important factor was whether or not a water quality test was judged nonconforming in the previous water quality test. It was confirmed that the temperature on the day of the inspection and the altitude of the mineral spring had an influence on whether the water quality was unsuitable.

• The Journal of the Korea Contents Association
• /
• v.12 no.10
• /
• pp.1-10
• /
• 2012

Many statistical methods have been adapted for terminology recognition to improve its accuracy. However, since previous studies have been carried out in a single core or a single machine, they have difficulties in real-time analysing explosively increasing documents. In this study, the task where bottlenecks occur in the process of terminology recognition is classified into linguistic processing in the process of 'candidate terminology extraction' and collection of statistical information in the process of 'terminology weight assignment'. A terminology recognition system is implemented and experimented to address each task by means of the distributed parallel processing-based MapReduce. The experiments were performed in two ways; the first experiment result revealed that distributed parallel processing by means of 12 nodes improves processing speed by 11.27 times as compared to the case of using a single machine and the second experiment was carried out on 1) default environment, 2) multiple reducers, 3) combiner, and 4) the combination of 2)and 3), and the use of 3) showed the best performance. Our terminology recognition system contributes to speed up knowledge extraction of large scale science and technology documents.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.6B
• /
• pp.561-571
• /
• 2006

RFID technology, as one contact-less identification technology with using electronic tag, is the field where its multiple researches have been studied. Especially, in the field of supply chain management, when it is necessary to identify a lot of objects by real-time, performance of anti-collision algorithm for multiple tag identification has an effect on performance of entire system. GEN2 is global, open, interoperability protocol for UHF RFID system, the active researches and developments are processed about it. In this paper, we introduce an anti-collision algorithm based on Slotted-ALOHA for GEN2 protocol, focus on estimating number of tags and setting number of slots and review the previous works. We propose new scheme that can estimate number of tags efficiently, and define system efficiency newly when it's used in setting number of slots. We also present result of compare and analysis with previous scheme through simulation. The proposed scheme of estimating number of tags is easy to implement, number of tags which is less than 4 times of number of slots can estimate tags efficiently, we also proved that identification efficiency based on weighted-slot distinguished with definition of previous slot efficiency can improve system efficiency.

• Journal of Advanced Navigation Technology
• /
• v.15 no.4
• /
• pp.523-535
• /
• 2011

The precipitable water vapor (PW) was estimated using Global Navigation Satellite System (GNSS) from several GNSS stations within the Korean Peninsula. Nearby radiosonde sites covering the GNSS stations were used for the comparison and validation of test results. GNSS data recorded under typical and severe weather conditions were used to generalize our approach. Based on the analysis, we have confirmed that the derived PW values from the GNSS observables were well agreed on the estimates from the radiosonde observables within 10 mm level. Assuming that the GNSS observables could be a good weather monitoring tool, we further tested the performance of the current WAAS tropospheric delay model, UNB3, in the Korean Peninsula. Especially, the wet zenith delays estimated from the GNSS observables and from UNB3 delay model were compared. Test results showed that the modelled approach for the troposphere (i.e., UNB3) did not perform well especially under the wet weather conditions in the Korean Peninsula. It was suggested that a new model or a near real-time model (e.g., based on regional model from GNSS or numerical weather model) would be highly desirable for the Korean WA-DGNSS to minimize the effects of the tropospheric delay and hence to achieve high precision vertical navigation solutions.

• The Journal of the Acoustical Society of Korea
• /
• v.32 no.2
• /
• pp.95-103
• /
• 2013

An underwater acoustic (UWA) communication in shallow water is strongly affected by the water surface and the seabed acoustical properties. Every reflected signal to receiver experiences a time-variant scattering in sea surface roughness and a grazing-angle-dependent reflection loss in bottom. Consequently, the performance of UWA communication systems is degraded, and high-speed digital communication is disrupted. If there is a dominant signal path such as a direct path, the received signal is modeled statistically as Rice fading but if not, it is modeled as Rayleigh fading. However, it has been known to be very difficult to reproduce the statistical estimation by real experimental evaluation in the sea. To give an insight for this scattering and grazing-angle-dependent bottom reflection loss effect in UWA communication, authors conduct experiments to quantify these effects. The image is transmitted using binary frequency shift keying (BFSK) modulation. The quality of the received image is shown to be affected by water surface scattering and grazing-angle-dependent bottom reflection loss. The analysis is based on the transmitter to receiver range and the receiver depth dependent image quality and bit error rate (BER). The results show that the received image quality is highly dependent on the transmitter-receiver range and receiver depth which characterizes the channel coherence bandwidth.

• Journal of the Institute of Convergence Signal Processing
• /
• v.9 no.1
• /
• pp.24-30
• /
• 2008

In this paper we analyze the performance of an adaptive sector cell system, which is adopted to maintain the traffic balance between sectors and to utilize the cell resources effectively, using the data collected from real channel environments. In the data measurements, we transmitted the QPSK modulated signal with carrier frequency of 1.95GHz and received the signals using the 8x4 array antenna equipped on the top of buildings in the urban area. We analyzed the angular distribution and the delay spread of a user signal and analyzed angular distribution of mobile users in a cell using the collected data. Also, we propose the vector channel modeling using the estimated pdf(probability distribution function) of the analyzing results. Through the proposed channel modeling the improvement of the call blocking rate was analyzed when using the adaptive sector cell system, and computer simulations show that the call blocking rate of the adaptive sector cell system was much lower than that of the fixed sector cell system. Additionally, it shows that the call blocking rate increases severely in the fixed sector cell system while the difference of the call blocking rate was smaller in the adaptive sector cell system, as the user density of the spatial distribution increases.

• Journal of the Korea Institute of Building Construction
• /
• v.10 no.1
• /
• pp.199-212
• /
• 2010

To establish an infrastructure for technology and information in the domestic construction industry, several construction regulations pertaining to construction information have been institutionalized. However, there are major problems with the domestic information classification system, earned value management (EVM) and project management information system (PMIS). In particular, the functions of the current PMIS have consisted of a builder-oriented system, and as EVM is not applied to PMIS, the functions of reporting, analysis and forecast for owners are lacking. Moreover, owners cannot confirm information on construction schedule and cost in real time due to the differences between the EVM and PMIS operation systems. The purpose of this study is to provide a framework that is capable of operating PMIS efficiently under an e-business environment, by providing a proposal on how to establish a work breakdown structure (WBS) and an EVM - PMIS integration model, so that PMIS may provide the function of EVM, and stakeholders may have all information in common. At the core of EVM - PMIS integration is the idea that EVM and PMIS have the same operation system, in order to be an activity-based system. The principle of the integration is data integration, in which the information field of an activity is connected with the field of a relational database table consisting of sub-modules for the schedule and cost management function of PMIS using a relational database management system. Therefore, the planned value (PV), cost value (CV), actual cost (AC), schedule variance (SV), schedule performance index (SPI), cost variance (CV) and cost performance index (CPI) of an activity are connected with the field of the relational database table for the schedule and cost sub-modules of PMIS.

• Journal of the Korean Solar Energy Society
• /
• v.30 no.6
• /
• pp.73-80
• /
• 2010

This study was conducted to estimate the relative performance of modules with changed characteristics due to long term exposure to the outdoor environment, with a specially made test device for simultaneous measurement of real time power output from the photovoltaic array, taking into account the inclined panel, direct irradiation, power being generated, temperature as well as the optimal analysis timing. In terminology description, M is an abbreviation of module and Group A, Group B are 10 modules series connection (1~10 of M), (11~20 of M) for each of them respectively. The overall mean voltage difference of M-18 with the lowest power output and M-14 with the highest output is-2.13V and it was identifiable that voltage difference was more concentrated to Group B. In addition, in case of M-2 and M-7, M-8, when compared with M-14, the overall mean voltage difference was -0.92V, -1.56 and -0.91V respectively showing the more concentration to Group A. When the temperature of module went up by $1^{\circ}C$, the mean voltage was reduced by 0.35V. For current, Group A was lower than Group B by-0.022A and the ratio of each group was 49.68% and 50.32% respectively, presumably the module with deteriorated properties were more concentrated to Group A relatively. From the comparison of relations with the comprehensive accumulation, M-2, M-7, M-8, M-16 and M-18 were those with deterioration of performance to the worst, thereby requiring precision examination. In comparative efficiency, M-14 was the most excellent one as 12.19% while M-18 as 10.53% was identified that its efficiency was comparatively rapidly reduced.

• Journal of Navigation and Port Research
• /
• v.40 no.3
• /
• pp.159-164
• /
• 2016

The goal of this study is to implement a communication system that can monitor the status of the nacelle using the power cable itself, without the dedicated communication lines such as an UTP cable and optical fiber for the offshore wind turbine. An inductive coupling powerline communication system for a MW class offshore wind turbine was proposed and its communication performance was demonstrated. The inductive couplers was designed for operation at up to 500 A using a ferrite composite materials. Field test was carried out on the wind farms of Jeju island. Using the iperf communication test program, we have obtained more than 15 Mbps data transmission rate through the 100 m power cable that was installed between the nacelle and the bottom of the power converter. In the data transmission stability test for a week, there was no failure ever. The minimum transmission rate was 15 Mbps and the average data rate was about 20 Mbps. Next, we have installed an infrared camera inside the nacelle in order to measure the temperature distribution and variation of the nacelle. The real-time thermal image taken by the camera was successfully sent to the monitoring system without error.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.5
• /
• pp.323-330
• /
• 2023

The development of an analysis model that reflects the microstructure characteristics of polyvinyl alcohol (PVA) fiber-reinforced cementitious composites, which have a highly complex microstructure, enables synergy between efficient material design and real experiments. PVA fiber orientations are an important factor that influences the mechanical behavior of PVA fiber-reinforced cementitious composites. Owing to the difficulty in distinguishing the gray level value obtained from micro-CT images of PVA fibers from adjacent phases, fiber segmentation is time-consuming work. In this study, a micro-CT test with a voxel size of 0.65 ㎛³ was performed to investigate the three-dimensional distribution of fibers. To segment the fibers and generate training data, histogram, morphology, and gradient-based phase-segmentation methods were used. A U-net model was proposed to segment fibers from micro-CT images of PVA fiber-reinforced cementitious composites. Data augmentation was applied to increase the accuracy of the training, using a total of 1024 images as training data. The performance of the model was evaluated using accuracy, precision, recall, and F1 score. The trained model achieved a high fiber segmentation performance and efficiency, and the approach can be applied to other specimens as well.