• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.3
• /
• pp.95-100
• /
• 2004

A control alogorithm using sliding mode and neural network for Buck type DC-DC converter is presented. Also, we conform a rightness the proposal method by comparing a theoretical values and experimental values obtained from experiment using DSP(digital signal processor). Performance comparisons made with the general hysteresis controller clearly bring out the superior performance of the proposal neural network controller. This paper will be applied to other power conversion system using the proposal neural network controller.

• Smart Media Journal
• /
• v.4 no.1
• /
• pp.25-32
• /
• 2015

Powering wireless sensors with energy harvested from the environment is coming of age due to the increasing power densities of both storage and harvesting devices and the electronics performing energy efficient energy conversion. In order to maximize the functionality of the wireless sensor network, minimize missing packets, minimize latency and prevent the waste of energy, problems like congestion and inefficient energy usage must be addressed. Many sleep-awake protocols and efficient message priority techniques have been developed to properly manage the energy of the nodes and to minimize congestion. For a WSN that is operating in a strictly energy constrained environment, an energy-efficient transmission strategy is necessary. In this paper, we present a novel transmission priority decision scheme for a heterogeneous body sensor network composed of normal nodes and an energy harvesting node that acts as a cluster head. The energy harvesting node's decision whether or not to clear a normal node for sending is based on a set of metrics which includes the energy harvesting node's remaining energy, the total harvested energy, the type of message in a normal node's queue and finally, the implementation context of the wireless sensor network.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.7 no.4
• /
• pp.212-222
• /
• 2014

In the current study, the authors propose a method for extracting license plate regions by means of a neural network trained to output the plate's center of gravity. The method is shown to be effective. Since the learning pattern presentation positions are defined by random numbers, a different pattern is submitted to the neural network for learning each time, which enables it to form a neural network with high universality of coverage. The article discusses issues of the optimal learning surface for a license plate covered by the learning pattern, the effect of suppression learning of the number and pattern enlargement/reduction and of concentration value conversion. Results of evaluation tests based on pictures of 595 vehicles taken at an under-ground parking garage demonstrated detection rates of 98.5%, 98.7%, and 100%, respectively.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.9
• /
• pp.82-92
• /
• 1998

An international standard for the product model data, STEP, and a standard for the distributed object technology, CORBA, will play a very important role in the future manufacturing environment. These two technologies provide background for the sharing of product data and the integration of applications on the network. This paper describes a prototype CAD/CAE environment that is integrated on the network by STEP and CORBA. Several application servers and client software were developed to verify the proposed concept. The present CAD/CAE environments are composed of several individual software components which are not tightly integrated. They also do not utilize the rapidly expanding network and object technologies for the collaboration in the product design process. In the design process in a large organization, sharing of application resources, design data and analysis data through the network will greatly enhance the productivity. The integration between applications can be supported by two key technologies, CORBA(Common Object Request Broker Architecture) and STEP(Standard for the Exchange of Product Model Bata). The CORBA provides interoperability between applications on different machines in heterogeneous distributed environments and seamlessly interconnects distributed object systems. Moreover, if all the data in the CAD/CAE environment are based on the STEP, then we can exclude all the data conversion problems between the application systems.

• The Journal of the Convergence on Culture Technology
• /
• v.5 no.3
• /
• pp.165-170
• /
• 2019

This study examined the meaning of employees' university-graduate school experiences for job conversion. For it, date was collected and analyzed by in-depth interview for 7 employees with university-graduate school experiences from Jun. to Doc. 2018. As a result, 3 great ranges and 8 sub-factors of the meaning of employees' university-graduate school experience for job conversion. were drawn. First is the change of value changing new view and recognition. Second is confidence improvement, motivation, and psychological change recovering interests and aptitudes. Third is changes of actual life that obtain pleasure of leaning, formation of personal network, and family supports. From the above results, employees' university-graduate school experience may let them challenge new works with confidence and be analyzed as appearance of identity finding the work suitable for their aptitude and living a subjective life.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.81-86
• /
• 2024

The Image Signal Processor (ISP) converts RAW images captured by the camera sensor into user-preferred sRGB images. While RAW images contain more meaningful information for image processing than sRGB images, RAW images are rarely shared due to their large sizes. Moreover, the actual ISP process of a camera is not disclosed, making it difficult to model the inverse process. Consequently, research on learning the conversion between sRGB and RAW has been conducted. Recently, the ParamISP[1] model, which directly incorporates camera parameters (exposure time, sensitivity, aperture size, and focal length) to mimic the operations of a real camera ISP, has been proposed by advancing the simple network structures. However, existing studies, including ParamISP[1], have limitations in modeling the camera ISP as they do not consider the degradation caused by lens shading, optical aberration, and lens distortion, which limits the restoration performance. This study introduces Positional Encoding to enable the camera ISP neural network to better handle degradations caused by lens. The proposed positional encoding method is suitable for camera ISP neural networks that learn by dividing the image into patches. By reflecting the spatial context of the image, it allows for more precise image restoration compared to existing models.

• Smart Media Journal
• /
• v.11 no.3
• /
• pp.26-30
• /
• 2022

Spiking neural network is a neural network that applies the working principle of real brain neurons. Due to the biological mechanism of neurons, it consumes less power for training and reasoning than conventional neural networks. Recently, as deep learning models become huge and operating costs increase exponentially, the spiking neural network is attracting attention as a third-generation neural network that connects convolution neural networks and recurrent neural networks, and related research is being actively conducted. However, in order to apply the spiking neural network model to the industry, a lot of research still needs to be done, and the problem of model retraining to apply a new model must also be solved. In this paper, we propose a method to minimize the cost of model retraining by extracting the weights of the existing trained deep learning model and converting them into the weights of the spiking neural network model. In addition, it was found that weight conversion worked correctly by comparing the results of inference using the converted weights with the results of the existing model.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.10 no.1
• /
• pp.124-130
• /
• 2011

In this paper, we propose a method for the efficient power use in power outage. The cost is high and conversion into the other function is difficult since the terminal used in conventional home network system has a dedicated function. This method is based on home network system employed with photovoltaic system. Throughout the modularization of the terminal, the variable function and cost reduction can be achieved. Additionally the unit cost can be reduced by using the power line communication. The prototype of light-control terminal has been implemented. The experimental results with the terminal show the performance of light-control system and the possibility of the commercialization can be achieved.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.9 no.3
• /
• pp.9-16
• /
• 1972

In this article it is shown that a GIC circuit with conversion "s" can be realized and a inductor is realized as a RC active two terminal network by using it instead of a gyrator. It is also shown that arbitary stable transfer functions can be realized as the open-circuit voltage ratio of 2 port networks which include GIC;s. In relizing the GIC circuit using transistors, it is made clear that the nullatornorator model of atransistor can be successfully applied at least in the frequency range below 10kHz. The synthesis method using GIC's is characterized with the followings; First, arbitrary stable transfer functions are realized systematically by repeating very simple network structure. Second, in the overall network all circuit elements except GICs are only resistors. Third, the number of condensers in the overall network necessary for realizing the transfer function of n-th order are n, which is believed to be the least number expected. expected.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.24 no.9 s.180
• /
• pp.2361-2370
• /
• 2000

The introduction of inexpensive cylinder pressure sensors provides new opportunities for precise engine control. This paper presents a control strategy of spark advance based upon cylinder pressure of spark ignition engines. A location of peak pressure(LPP) is the major parameter for controlling the spark timing, and also the UP is estimated, using a multi-layer feedforward neural network, which needs only five pressure sensor output voltage samples at -40˚, -20˚, 0˚, 20˚, 40˚ after top dead center. The neural network plays an important role in mitigating the A/D conversion load of an electronic engine controller by increasing the sampling interval from 10 crank angle(CA) to 20˚ CA. A proposed control algorithm does not need a sensor calibration and pegging(bias calculation) procedure because the neural network estimates the UP from the raw sensor output voltage. The estimated LPP can be regarded as a good index for combustion phasing, and can also be used as an MBT control parameter. The feasibility of this methodology is closely examined through steady and transient engine operations to control individual cylinder spark advance. The experimental results have revealed a favorable agreement of individual cylinder optimal combustion phasing.