• International journal of advanced smart convergence
• /
• v.8 no.1
• /
• pp.141-146
• /
• 2019

These days, there are many applications using neural networks as parts of their system. On the other hand, adversarial examples have become an important issue concerining the security of neural networks. A classifier in neural networks can be fooled and make it miss-classified by adversarial examples. There are many research to encounter adversarial examples by using denoising methods. Some of them using GAN (Generative Adversarial Network) in order to remove adversarial noise from input images. By producing an image from generator network that is close enough to the original clean image, the adversarial examples effects can be reduced. However, there is a chance when adversarial noise can survive the approximation process because it is not like a normal noise. In this chance, we propose a research that utilizes high-level representation in the classifier by combining GAN network with a trained U-Net network. This approach focuses on minimizing the loss function on high representation terms, in order to minimize the difference between the high representation level of the clean data and the approximated output of the noisy data in the training dataset. Furthermore, the generated output is checked whether it shows minimum error compared to true label or not. U-Net network is trained with true label to make sure the generated output gives minimum error in the end. At last, the remaining adversarial noise that still exist after low-level approximation can be removed with the U-Net, because of the minimization on high representation terms.

• ETRI Journal
• /
• v.42 no.6
• /
• pp.951-964
• /
• 2020

This paper presents a lightweight true random number generator (TRNG) using beta radiation that is useful for Internet of Things (IoT) security. In general, a random number generator (RNG) is required for all secure communication devices because random numbers are needed to generate encryption keys. Most RNGs are computer algorithms and use physical noise as their seed. However, it is difficult to obtain physical noise in small IoT devices. Since IoT security functions are required in almost all countries, IoT devices must be equipped with security algorithms that can pass the cryptographic module validation programs of each country. In this regard, it is very cumbersome to embed security algorithms, random number generation algorithms, and even physical noise sources in small IoT devices. Therefore, this paper introduces a lightweight TRNG comprising a thin-film beta-radiation source and integrated circuits (ICs). Although the ICs are currently being designed, the IC design was functionally verified at the board level. Our random numbers are output from a verification board and tested according to National Institute of Standards and Technology standards.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.619-628
• /
• 2019

In this paper, we designed a beta ray sensor for a true random number generator. Instead of biasing the gate of the PMOS feedback transistor to a DC voltage, the current flowing through the PMOS feedback transistor is mirrored through a current bias circuit designed to be insensitive to PVT fluctuations, thereby minimizing fluctuations in the signal voltage of the CSA. In addition, by using the constant current supplied by the BGR (Bandgap Reference) circuit, the signal voltage is charged to the VCOM voltage level, thereby reducing the change in charge time to enable high-speed sensing. The beta ray sensor designed with 0.18㎛ CMOS process shows that the minimum signal voltage and maximum signal voltage of the CSA circuit which are resulted from corner simulation are 205mV and 303mV, respectively. and the minimum and maximum widths of the pulses generated by comparing the output signal through the pulse shaper with the threshold voltage (VTHR) voltage of the comparator, were 0.592㎲ and 1.247㎲, respectively. resulting in high-speed detection of 100kHz. Thus, it is designed to count up to 100 kilo pulses per second.

• (The)Study of the Eastern Classic
• /
• no.36
• /
• pp.85-104
• /
• 2009

This study examines a study of communication factor in Lunyu, as a communication text, in terms of communicator, audience, message, communication factor, communication text, interdisciplinary research. In many respects, it is generally accepted that Lunyu have been the generic references of the Oriental culture. Lunyu consider ethics, logic, and practicability as the qualifying requirement of communicator, asserting that communicator should speak true language, like a "chuntzu"(君子) does, and should also put their language into practice. The audience's attitude and method as contained in Lunyu are that hearers should have sharp ears for language, hear selectively the right language, and use the language suitable to the situation. It is also emphasized that the Hearer should actively lead in the situation of transactional communications. In Lunyu, one property of message is that language, which determines the rise and fall of a nation and is also the basis of judgement for other people, should comply with ethics and reasons and sould also be put into practice. In other words, credible message, as the practice of language, is the practical requirement of ethics and the qualification of a "chuntzu"(君子, superior man) in ruling the nation or conducting one's life.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.8
• /
• pp.623-628
• /
• 2000

Video-on-demand system in which users can request any video through the network at any time is made possible by rapid increase in network bandwidth and capacity of the media server. However true video-on-demand system cannot support all requests since bandwidth requirement is still too demanding. Therefore efficient bandwidth reduction algorithm is necessary. both the piggybacking method and the batching method are novel solutions that can provide more logical number of streams than the physical system can support. Of course each of them has its pros and cons. hence piggybacking with batching-by-size can take advantage of both the schemes. Some parameters such as the size of batch and the size of the catch-up window should be adjusted and order to maximize the bandwidth reduction for piggybacking with batching-by-size method. One of the most important parameters is decided optimally in this paper. Simulation shows that the optimized parameter can achieve considerable reductionand consequently remarkable enhancement in performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.4
• /
• pp.1931-1953
• /
• 2017

The reverse skyline query plays an important role in information searching applications. This paper deals with continuous reverse skyline queries in sensor networks, which retrieves reverse skylines as well as the set of nodes that reported them for continuous sampling epochs. Designing an energy-efficient approach to answer continuous reverse skyline queries is non-trivial because the reverse skyline query is not decomposable and a huge number of unqualified nodes need to report their sensor readings. In this paper, we develop a new algorithm that avoids transmission of updates from nodes that cannot influence the reverse skyline. We propose a data mapping scheme to estimate sensor readings and determine their dominance relationships without having to know the true values. We also theoretically analyze the properties for reverse skyline computation, and propose efficient pruning techniques while guaranteeing the correctness of the answer. An extensive experimental evaluation demonstrates the efficiency of our approach.

• Journal of Biomedical Engineering Research
• /
• v.30 no.3
• /
• pp.185-190
• /
• 2009

Biomedical mobile devices for ubiquitous healthcare consist of biomedical sensors and communication terminal. They have two types of configuration. One is the sensor-network type device using wired or wireless communication with intelligent sensors to acquire biomedical data. The other is the sensor embedded type device, where the data can be acquired directly by itself. There are many examples of sensor network type, such as, fall detection sensor, blood glucose sensor, and ECG sensors networked with commercial PDA phone and commercial phone terminal for ubiquitous healthcare. On the other hand, sensor embedded type mounts blood glucose sensor, accelerometer, and etc. on commercial phone. However, to enable true ubiquitous healthcare, motion sensing is essential, because users go around anywhere and their signals should be measured and monitored, when they are affected by the motion. Therefore, in this paper, two biomedical mobile devices with motion monitoring function were addressed. One is sensor-network type with motion monitoring function, which uses Zigbee communication to measure the ECG, PPG and acceleration. The other is sensor-embedded type with motion monitoring function, which also can measure the data and uses the built-in cellular phone network modem for remote connection. These devices are expected to be useful for ubiquitous healthcare in coming aged society in Korea.

• ETRI Journal
• /
• v.28 no.5
• /
• pp.684-687
• /
• 2006

In this letter, we evaluate the system performance of a space-time block coded (STBC) multicarrier (MC) DS-CDMA system over a time selective fading channel, with imperfect channel knowledge. The average bit error rate impairment due to imperfect channel information is investigated by taking into account the effect of the STBC position. We consider two schemes: STBC after spreading and STBC before spreading in the MC DS-CDMA system. In the scheme with STBC after spreading, STBC is performed at the chip level; in the scheme with STBC before spreading, STBC is performed at the symbol level. We found that these two schemes have various channel estimation errors, and that the system with STBC before spreading is more sensitive to channel estimation than the system with STBC after spreading. Furthermore, derived results prove that a high spreading factor (SF) in the MC DS-CDMA system with STBC before spreading leads to high channel estimation error, whereas for a system with STBC after spreading this statement is not true.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.28 no.12
• /
• pp.90-94
• /
• 2014

Recently, various types of trains are being operated in the same railroad whose designated stop stations are not the same. In other words, each train does not stop at every station. This could cause the pass-through accidents in which a train passes a station where it is supposed to stop. Although the number of train accidents is on the decrease, this type of error is still an important problem to be resolved. It is true that the burdens of train drivers are growing because the number of high speed trains and non-stop stations is increasing in the current railway system. However, another major reason for this accident is that there is no prevention system for these errors in the current train control system although it is being constantly improved. The purpose of this study is to reinforce the train stop system with wireless communication equipment in order to decrease this type of accident and to maximize the safety of railway transportation.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.6
• /
• pp.405-416
• /
• 2019

Due to the development of information and communication technology, the production and processing speed of data is getting faster. To classify objects using machine learning, which is a field of artificial intelligence, data required for training can be easily collected due to the development of internet and geospatial information technology. In the field of geospatial information, machine learning is also being applied to classify or recognize objects using images and point clouds. In this study, the problem of manually constructing training data using existing digital map version 1.0 was improved, and the technique of classifying roads, buildings and vegetation using image and point clouds were proposed. Through experiments, it was possible to classify roads, buildings, and vegetation that could clearly distinguish colors when using true ortho-image with only RGB (Red, Green, Blue) bands. However, if the colors of the objects to be classified are similar, it was possible to identify the limitations of poor classification of the objects. To improve the limitations, random forest and support vector machine techniques were applied after band fusion of true ortho-image and normalized digital surface model, and roads, buildings, and vegetation were classified with more than 85% accuracy.