• The Journal of Korea Robotics Society
• /
• v.14 no.1
• /
• pp.40-49
• /
• 2019

Reinforcement learning has been applied to various problems in robotics. However, it was still hard to train complex robotic manipulation tasks since there is a few models which can be applicable to general tasks. Such general models require a lot of training episodes. In these reasons, deep neural networks which have shown to be good function approximators have not been actively used for robot manipulation task. Recently, some of these challenges are solved by a set of methods, such as Guided Policy Search, which guide or limit search directions while training of a deep neural network based policy model. These frameworks are already applied to a humanoid robot, PR2. However, in robotics, it is not trivial to adjust existing algorithms designed for one robot to another robot. In this paper, we present our implementation of Guided Policy Search to the robotic arms of the Baxter Research Robot. To meet the goals and needs of the project, we build on an existing implementation of Baxter Agent class for the Guided Policy Search algorithm code using the built-in Python interface. This work is expected to play an important role in popularizing robot manipulation reinforcement learning methods on cost-effective robot platforms.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.6
• /
• pp.1223-1228
• /
• 2018

LTE-R(Long Term Evolution - Railway), as a mobile communication technology to support safety operation of train on high speed railway system, have to be equipped appropriately by hand-over method for high moving environment of train. In this paper, hand-over performance of LTE-R network, which is used in railway communication system, is analyzed and hand-over implementation condition is suggested with this analysis. The performance analysis of this study is carried out with computer simulation with NS(Network Simulator)-3. SINR(Signal to Interference and Noise Ratio) and RSRP(Reference Signal Received Power), and Data Rate is used for hand-over performance parameter.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.578-581
• /
• 2014

Blackhole attack, a kinds of attacks to routing function, can cause critical effects to network transmission function, Especially, on MANET(Mobile Ad-hoc Network) which it is not easy to prepare functions to respond malicious intrusion, transmission functions of entire networks could be degraded. In this paper, effects of blackhole attack to network transmission performance is analyzed on lattice structured MANET. Specially, performance is measured for various location of blackhole attack on lattice MANET, and compared with the performance of random structured MANET. This paper is done with computer simulation, VoIP(Voice over Internet Protocol) traffic is used in simulation. The results of this paper can be used for data to deal with blackhole attack.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.217-220
• /
• 2014

Wireless sensor network（WSN） consists by a large number of low-cost micro-sensor nodes, collaborate to achieve the perception of information collection, processing and transmission tasks in deployment area. It can be widely used in national defense, intelligent transportation, medical care, environmental monitoring, precision agriculture, and industrial automation and many other areas. One of the key technologies of sensor networks is the data maintenance management technology. In this paper we analyze the data management technology of wireless sensor network and pointed their problems.

• Journal of Korea Multimedia Society
• /
• v.22 no.9
• /
• pp.1000-1010
• /
• 2019

Image semantic segmentation and dehazing are key tasks in the computer vision. In recent years, researches in both tasks have achieved substantial improvements in performance with the development of Convolutional Neural Network (CNN). However, most of the previous works for semantic segmentation assume the images are captured in clear weather and show degraded performance under hazy images with low contrast and faded color. Meanwhile, dehazing aims to recover clear image given observed hazy image, which is an ill-posed problem and can be alleviated with additional information about the image. In this work, we propose a deep multi-task network for simultaneous semantic segmentation and dehazing. The proposed network takes single haze image as input and predicts dense semantic segmentation map and clear image. The visual information getting refined during the dehazing process can help the recognition task of semantic segmentation. On the other hand, semantic features obtained during the semantic segmentation process can provide cues for color priors for objects, which can help dehazing process. Experimental results demonstrate the effectiveness of the proposed multi-task approach, showing improved performance compared to the separate networks.

• Journal of Multimedia Information System
• /
• v.6 no.3
• /
• pp.131-138
• /
• 2019

With the rapid evolution of communication technology, it became possible to overcome the spatial and temporal limitations faced by humans to some extent. Furthermore, the quality of personal life was revolutionized with the emergence of the personal communication device commonly known as the smart phone. In terms of defense networks, however, due to restrictions from the military and security perspectives, the use of smart phones has been prohibited and controlled in the army; thus, they are not being used for any defense strategy purposes as yet. Despite the current consideration of smart phones for military communication, due to the difficulties of network configuration and the high cost of the necessary communication devices, the main tools of communication between soldiers are limited to the use of flag, voice or hand signals, which are all very primitive. Although these primitive tools can be very effective in certain cases, they cannot overcome temporal and spatial limitations. Likewise, depending on the level of the communication skills of each individual, communication efficiency can vary significantly. As the term of military service continues to be shortened, however, types of communication of varying efficiency depending on the levels of skills of each individual newly added to the military is not desirable at all. To address this problem, it is essential to prepare an intuitive network configuration that facilitates use by soldiers in a short period of time by easily configuring the strategy network at a low cost while maintaining its security. Therefore, in this article, the author proposes a Zigbee-based local strategic network by using Opnet and performs a simulation accordingly.

• Journal of Internet Computing and Services
• /
• v.20 no.1
• /
• pp.11-16
• /
• 2019

Wireless sensor network is important for precision farming to monitor the growth environment of crops in open field, but radio signals are susceptible to different types of interference such as weather and physical objects. This paper designs and implements an environmental monitoring and weather forecast acquisition systems for smart field irrigation based on LoRa(Long Range) and then applies it to a test bed. And we evaluate the network reliability in terms of packet transmission success rate by comparing its condition on two criteria; the existence of obstacle or rain. The results show that much rain falls can affect on packet loss in LoRa field networks with obstacles.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.2
• /
• pp.103-109
• /
• 2019

Due to the development of ICT technology, the IoT environment for connecting objects in the vicinity to networks and utilizing information about objects in various fields is getting attention, and security threats are also increasing. In order to solve the increasing security problem in IoT environment, we are studying methods that use certificate, encryption, hash calculation and block chain in the private sector. However, the security authentication method which overcomes the performance gap between devices and has compatibility with various devices It has not been proposed yet. In this paper, we propose an authentication method that can achieve wide compatibility by minimizing the influence of IoT device environment.

• Journal of information and communication convergence engineering
• /
• v.16 no.4
• /
• pp.252-257
• /
• 2018

A wideband double radiator circular disc annular monopole antenna is proposed is this work. The radiators are etched on the surfaces of two Taconic TLY-5 substrates with a circular hole cut out of each of the radiators initially at the centers of the radiators with subsequent downward displacement of the holes. The antenna is designed with a two-step feeding transformer system for impedance matching between the input power source supplied by a $50-{\Omega}$ SMA connector and the monopole radiators. The transformer system improves the bandwidth performance at higher frequencies. The proposed antenna achieves a wideband having the capability of working between 0.645 and 18.775 GHz, corresponding to a -10 dB bandwidth of 186.7% with gain ranging from 0.95 to 8.26 dBi. In comparison to other metal disc planar monopole antennas, the proposed antenna has a small total size width due to the size of the ground plane, which has a diameter 100 mm. The frequency range of the antenna provides applications in global positioning systems, mobile communications, ultra-wideband short distance communications, and wireless computer networks.

• Convergence Security Journal
• /
• v.21 no.3
• /
• pp.3-11
• /
• 2021

Most of the internet security protocols rely on classical algorithms based on the mathematical complexity of the integer factorization problem, which becomes vulnerable to a quantum computer. Recent progresses of quantum computing technologies have highlighted the need for applying quantum key distribution (QKD) on existing network protocols. We report the development and integration of a plug & play QKD device with a commercial IPSec device by replacing the session keys used in IPSec protocol with the quantum ones. We expect that this work paves the way for enhancing security of the star-type networks by implementing QKD with the end-to-end IP communication.