• Journal of Broadcast Engineering
• /
• v.23 no.6
• /
• pp.855-865
• /
• 2018

Sound event detection is one of the research areas to model human auditory cognitive characteristics by recognizing events in an environment with multiple acoustic events and determining the onset and offset time for each event. DCASE, a research group on acoustic scene classification and sound event detection, is proceeding challenges to encourage participation of researchers and to activate sound event detection research. However, the size of the dataset provided by the DCASE Challenge is relatively small compared to ImageNet, which is a representative dataset for visual object recognition, and there are not many open sources for the acoustic dataset. In this study, the sound events that can occur in indoor and outdoor are collected on a larger scale and annotated for dataset construction. Furthermore, to improve the performance of the sound event detection task, we developed a dual CNN structured sound event detection system by adding a supplementary neural network to a convolutional neural network to determine the presence of sound events. Finally, we conducted a comparative experiment with both baseline systems of the DCASE 2016 and 2017.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.6
• /
• pp.483-488
• /
• 2018

Far field input utterance is one of the major causes of performance degradation of speaker verification systems. In this study, we used teacher student learning framework to compensate for the performance degradation caused by far field utterances. Teacher student learning refers to training the student deep neural network in possible performance degradation condition using the teacher deep neural network trained without such condition. In this study, we use the teacher network trained with near distance utterances to train the student network with far distance utterances. However, through experiments, it was found that performance of near distance utterances were deteriorated. To avoid such phenomenon, we proposed techniques that use trained teacher network as initialization of student network and training the student network using both near and far field utterances. Experiments were conducted using deep neural networks that input raw waveforms of 4-channel utterances recorded in both near and far distance. Results show the equal error rate of near and far-field utterances respectively, 2.55 % / 2.8 % without teacher student learning, 9.75 % / 1.8 % for conventional teacher student learning, and 2.5 % / 2.7 % with proposed techniques.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.24 no.7
• /
• pp.953-959
• /
• 2018

This paper deals with controlling surge oscillations of a mooring vessel system under large external disturbances such as wind, waves and currents. A control synthesis based on Sliding Mode Control (SMC) with a Super-Twisting Algorithm (STA) has been applied to suppress nonlinear surge oscillations of a two-point mooring system. Despite the advantages of robustness against parameter uncertainties and disturbances for SMC, chattering is the main drawback for implementing sliding mode controllers. First-order SMC shows convergence within the desired level of accuracy, in which chattering is the main obstacle related to the destructive phenomenon. Alternatively, STA completely eliminates chattering phenomenon with high accuracy even for large disturbances. SMC based on STA is an effective tool for the motion control of a nonlinear mooring system because it avoids the chattering problems of a first-order sliding mode controller. In addition, the error trajectories of controlled mooring systems implemented by means of STA form in the bounded region. Finally, the control gain effect of STA can be observed in sliding surface and position trajectory errors.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.8
• /
• pp.317-322
• /
• 2019

Programming judge systems for programming training support are typically built on the Web, where the examiners uploads a programming problem, which the student reads and submits an answer to the problem. The judge system executes the submitted answer of source code to provide feedback such as pass, failure, and error messages. Students who receive the feedback except for the pass continues debugging the source code until they are judged to pass. We developed an online judge system to support programming training and analyzed answers submitted by the students and found that many of the students who were not judged to pass that test did not know exactly where they were wrong but continued to solve the problem. The current judge system generally feeds runtime error messages back to students. However, with only runtime error message, it is difficult for student who train to find the wrong part of the answer. Therefore, in this paper, we propose a system that provides the feedback of programming failure by analyzing the control flow of the test cases used in the source code submitted by the student. The proposed system helps students find the wrong parts more quickly by feeding back the paths where faults in the control flow may exist. In addition, we show that this system is applicable to the answer source code that the actual student submitted.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.465-468
• /
• 2014

The image processing device sold by the market, which increases the comfort of the driver Around-View of the camera. This system while driving or when parked, came about to prevent accidents caused by driver error or disable the visibility of the system. However, it did not spread widely to the driver due to the problem of the high installation cost and complex installation process from the system for easy operation. Due to problems such as first, expensive equipment and second, the development environment is difficult and third, inconvenient installation process, it is not out because of the prohibitively high cost burden and difficult development environment, programmers and operators. I think if this is solved even one problem of this system would be able to access the user are a little more affordable. In this paper The AVM(Around-View Monitoring) system is proposed, the two problems that minimize expensive equipment, the installation process is inconvenient problem of the three aforementioned systems. Solved the problem caused by a lot of the cost by using low-cost USB device, and a rear camera. Was developed to facilitate the installation is possible by considering the inconvenient installation. Reducing the price paid by consumers because of the system.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.79-81
• /
• 2014

Recently, It is built various information systems evolve IT skills. But When you build the information system, Difficult to determine whether the appropriate scale and problems that rely heavily on SI companies and professionals. To solve this problem, Korea Information Security Agency, etc., based on the primary objective was to develop H/W Capacity Equation formally to each system type. But the problems are to present H/W capacity equation by discussion of the expert group of suppliers and relatively long that it is difficult to formally apply in the situation now so it is no longer the limit. In this study, we proposes proper capacity planning techniques, which can guarantee the best performance compared to the budget invested. For this purpose, we derived the proper H/W capacity equation by regression analysis to gather performance metrics and cost of various cases by simulation of a virtual environment in the cloud. Through this study, when capacity planning, It is possible to reduce costs that It is possible to build an information system based on the digitized data and build information system in an environment that does not rely on the SI business or professional.

• Journal of the Economic Geographical Society of Korea
• /
• v.21 no.4
• /
• pp.335-349
• /
• 2018

This study analyzes the effect of the changes in traffic environments such as transportation speeds on the time-distance accessibility for the public transportation passengers. To do this, we use passenger transaction databases of the Seoul metropolitan public transportation system: one week for each of the three years (2011, 2013, and 2015). These big data contain the information about time and space on the traffic trajectories of every passenger. In this study, the time-distances of links between subway stations and bus stops of the public transportation system at each time are calculated based on the actual travel time extracted from the traffic-card transaction database. The changes in the time-distance accessibility of the integrated transportation network from the experimental results can be summarized in two aspects. First, the accessibility tends to decline as the year goes by. This is because the transportation network becomes more complicated and then the average moving speed of the vehicles is lowered. Second, the accessibility tends to increase on the weekend in the analysis of accessibility changes by day. This tendency is because the bus speeds on bus routes on the weekend are faster than other days. In order to analyze the accessibility changes, we illustrate graphs of the vehicle speeds and the numbers of passengers by year and day.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.4
• /
• pp.406-412
• /
• 2019

In this paper, we propose two algorithms for classifying services with similar delay characteristics into three classes and allocating radio resources according to priority in LTE-A system where H2H and M2M services coexist. The first is to allocate resources from the higher priority class to the lower priority class, and each class gives priority to H2H over M2M, and the other is to give priority to H2H regardless of delay characteristics except for the class with the highest priority. The RA success probability was analyzed according to the access rate(${\alpha}$) of M2M devices in each class. In comparison with the conventional systems, it was improved from 0.5 to 0.52 for ${\alpha}_{2M}=0.05$ in two classes. In the three classes, the success probability was slightly increased from 0.5 to 0.57 for ${\alpha}_{2M}={\alpha}_{3M}=1$ and from 0.5 to 0.58 for ${\alpha}_{2M}=0.5$ and ${\alpha}_{3M}=0.1$. Although 6 services are considered in the proposed scheme, the RA success probability is almost similar to the previous scheme because the average arrival rate of H2H of each class is set to the same.

• The Journal of Society for e-Business Studies
• /
• v.24 no.2
• /
• pp.199-216
• /
• 2019

The public SW market is 3 trillion won, which is less than 10% of the total SW market. However, due to the nature of the domestic market, it is an important market with a relatively large impact on small and medium-sized software companies. In this market, government is operating the Public SW Project Demand Forecasting System in order to support the marketing activities of small and medium sized SW companies and establish a fair market order. The current system has limitations such as lack of user convenience, insufficient analysis capability and less business connection. This study was conducted to identify the problems of these systems and to propose a new system for improving the convenience of users and expanding the information utilization of SMEs. To this end, we analyzed the requirements of each stakeholder. We proposed the 2-phased forecasting cycle, the management cycle, and the system life cycle of public SW projects and created a unified identifier (UID) so that the information of those projects can be identified and linked among them. As a result, an integrated reference model of project information management based on system life cycle was developed, which can explain the demand forecasting and project information, and the improved processes was also designed to implement them. Through the result of this study, it is expected that integrated management of public SW projects will be possible.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.5
• /
• pp.499-506
• /
• 2019

This paper is concerned with the operating stability of 7kW inverter using SIC hybrid module and verifies the validity of the simulation results by comparing the result of the loss equation and the simulation result, Simulation results using Si module and SiC hybrid module are compared to compare switch loss and diode loss. Through the loss equation calculation, the conduction loss of SiC Hybrid module is 168W, switching loss is 9.3W, diode loss is 10.5nW, When compared with the simulation results, similar values were shown. As a result of comparing the simulation results of the Si module and the SiC Hybrid module, The total device loss of the Si module was 246.2W, and the total device loss of the SiC Hybrid module was 189.9W. The loss difference was 56.3W, which was about 0.8W. thereby verifying the reverse recovery characteristics of the SiC SBD. In addition, temperature saturation test was conducted to confirm the stability of SiC Hybrid module and Si module under high temperature saturation, In the case of the Si module, the output power was stopped at 4kW, and the SiC Hybrid module was confirmed to operate at 7kW. Based on this, an efficiency graph and a temperature graph are presented, and the Si module is graphed up to 4kW and the SiC Hybrid module is graphed up to 7kW.