• Journal of Korea Society of Digital Industry and Information Management
• /
• v.14 no.4
• /
• pp.17-26
• /
• 2018

Connected car plays an important role on Intelligent Transport System (ITS). ITS is able to secure drivers' convenience and safety, however, the overall system can be threatened by hacking attempt. Blockchain is one strong candidate of the remedy to promote the security of the ITS network. However, there will be many challenges to adopt previously proposed blockchain technologies to ITS. This work presents a new ITS structure based on blockchain technology. Proposed scheme includes three major layers. The first layer is central manager which is initiated once to register a certain connected car. The third layer is RSU (Road Side Unit) layer which exploits PoS (Proof of Stake) for consortium blockchains and retains real-time information. In addition, this layer performs block expiration based on timers to maintain manageable block length. In the second layer, the generated blocks of the third layer without expiration are housed as private blockchains. We finally demonstrate possible merits of newly proposed scheme.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.686-689
• /
• 2012

In this paper, the GaN-based LED characteristics are analyzed using ISE-TCAD. The LED consists of GaN barriers, active region of InGaN quantum well, AlGaN EBL(Electron Blocking Layer) and AlGaN HBL(Hole Blocking Layer) on GaN buffer layer. The output power characteristics of LED considering Auger recombination rate, thickness of quantum well and number of quantum wells are analyzed and some criteria for the design of LED are proposed.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.2
• /
• pp.699-707
• /
• 2015

Although digit character recognition has got a significant improvement in recent years, it is still challenging to achieve satisfied result if the data contains an amount of distracting factors. This paper proposes a novel digit character recognition approach using a multi-layer hierarchical model, Hybrid Restricted Boltzmann Machines (HRBMs), which allows the learning architecture to be robust to background distracting factors. The insight behind the proposed model is that useful high-level features appear more frequently than distracting factors during learning, thus the high-level features can be decompose into hybrid hierarchical structures by using only small label information. In order to extract robust and compact features, a stochastic 0-1 layer is employed, which enables the model's hidden nodes to independently capture the useful character features during training. Experiments on the variations of Mixed National Institute of Standards and Technology (MNIST) dataset show that improvements of the multi-layer hierarchical model can be achieved by the proposed method. Finally, the paper shows the proposed technique which is used in a real-world application, where it is able to identify digit characters under various complex background images.

• Journal of the Korean Graphic Arts Communication Society
• /
• v.22 no.2
• /
• pp.73-82
• /
• 2004

A thermal transfer method was developed novel method to form the phosphor screen for monochrom VPT. This method have advantages of simple process, clean environment, saving raw material and running-cost. But now applying phosphor screen for thermal transfer method has been formed three layers (phosphor layer, ITO layer and thermal adhesive layer) on the PET film as substrate. This is complex process, consumption of raw-material and require of high cost. Also ITO paste at present has been imported from Japan. To improve these problems, we have developed ITO paste as conductive paste by using ITO sol and binder resin (AA3003). Ito paste as developed in this study has both conductive and excellent thermal transfer abilities. Thus we could manufacture phosphor screen formed two layers (phosphor layer and ITO layer).

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.10a
• /
• pp.251-254
• /
• 2012

In this paper, the output characteristics of GaN-based LED considering quantum well structure are analyzed. The basic structure of the LED consists of active region of GaN barrier and InGaN quantum well between AlGaN EBL(Electron Blocking Layer) and AlGaN HBL(Hole Blocking Layer) on GaN buffer layer. The output power, internal quantum efficiency characteristics of LED active region considering thickness of quantum well, number of quantum well and doping of barrier are analyzed using ISE-TCAD.

• Journal of Information Processing Systems
• /
• v.15 no.3
• /
• pp.632-644
• /
• 2019

The remote monitoring and warning system for dangerous chemicals is designed with the concept of the Cyber-Physical System (CPS) in this paper. The real-time perception, dynamic control, and information service of major hazards chemicals are realized in this CPS system. The CPS system architecture, the physical layer and the applacation layer, are designed in this paper. The terminal node is mainly composed of the field collectors which complete the data acquisition of sensors and video in the physical layers, and the use of application layer makes CPS system safer and more reliable to monitor the hazardous chemicals. The cloud application layer completes the risk identification and the prediction of the major hazard sources. The early intelligent warning of the major dangerous chemicals is realized and the security risk images are given in the cloud application layer. With the CPS technology, the remote network of hazardous chemicals has been completed, and a major hazard monitoring and accident warning online system is formed. Through the experiment of the terminal node, it can be proved that the terminal node can complete the mass data collection and classify. With this experiment it can be obtained the CPS system is safe and effective. In order to verify feasible, the multi-risk warning based on CPS is simulated, and results show that the system solves the problem of hazardous chemicals enterprises safety management.

• ETRI Journal
• /
• v.41 no.2
• /
• pp.242-253
• /
• 2019

We devise a layer-wise hint training method to improve the existing hint-based knowledge distillation (KD) training approach, which is employed for knowledge transfer in a teacher-student framework using a residual network (ResNet). To achieve this objective, the proposed method first iteratively trains the student ResNet and incrementally employs hint-based information extracted from the pretrained teacher ResNet containing several hint and guided layers. Next, typical softening factor-based KD training is performed using the previously estimated hint-based information. We compare the recognition accuracy of the proposed approach with that of KD training without hints, hint-based KD training, and ResNet-based layer-wise pretraining using reliable datasets, including CIFAR-10, CIFAR-100, and MNIST. When using the selected multiple hint-based information items and their layer-wise transfer in the proposed method, the trained student ResNet more accurately reflects the pretrained teacher ResNet's rich information than the baseline training methods, for all the benchmark datasets we consider in this study.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.18 no.4
• /
• pp.13-21
• /
• 2022

Recently, in wireless video sensor networks(WVSN), various schemes for efficient video data transmission have been studied. In this paper, a layer based cooperative relaying(LCR) algorithm is proposed for minimizing scalable video transmission distortion from packet loss in WVSN. The proposed LCR algorithm consists of two modules. In the first step, a parameter based error propagation metric is proposed to predict the effect of each scalable layer on video quality degradation at low complexity. In the second step, a layer-based cooperative relay algorithm is proposed to minimize distortion due to packet loss using the proposed error propagation metric and channel information of the video sensor node and relay node. In the experiment, the proposed algorithm showed that the improvement of peak signal-to-noise ratio (PSNR) in various channel environments, compared to the previous algorithm(Energy based Cooperative Relaying, ECR) without considering the metric of error propagation.The proposed LCR algorithm minimizes video quality degradation from packet loss using both the channel information of relaying node and the amount of layer based error propagation in scalable video.

• Proceedings of the IEEK Conference
• /
• 2000.11a
• /
• pp.441-444
• /
• 2000

This paper covers the performance testing for layer 3 Ethernet switch based on various methodologies by which we can measure essential metrics such as throughput, latency, frame loss rate, and back to back frames. In the first place, layer 2 and layer 3 switch evolution is introduced followed by description of IP packet switching in layer 3 switch. And then, the above test metrics and test methodologies are illustrated as well. At last, we conduct the performance testing for layer 3 switch in case of transmitting packets of 64, 128, 256, 512, 1024, 1280, and 1518 byte size and analyze then results.

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.163-166
• /
• 2017

딥러닝은 모델이 복잡해질수록 Train 시간이 오래 걸리는 작업이다. Layer Normalization은 Train 시간을 줄이고, layer를 정규화 함으로써 성능을 개선할 수 있는 방법이다. 본 논문에서는 한국어 의미역 결정을 위해 Layer Normalization이 적용 된 Bidirectional LSTM CRF 모델을 제안한다. 실험 결과, Layer Normalization이 적용 된 Bidirectional LSTM CRF 모델은 한국어 의미역 결정 논항 인식 및 분류(AIC)에서 성능을 개선시켰다.