• Proceedings of the Korea Contents Association Conference
• /
• 2004.11a
• /
• pp.327-332
• /
• 2004

USB(Universal Serial Bus) is one of the most popular communication interfaces. When USB is used in an extended range, especially configurating In-home network by connecting multiple digital devices each other, USB interface uses the bandwidth in the way of TDM(Time Division Multiplexing) so that the bottleneck of bus bandwidth can be brought. In this paper, the more effective usage of bus bandwidth to overcome this situation is introduced. Basically, in order to realize the system for transferring realtime moving picture data among digital information devices, we analyze USB transfer types and Descriptors and introduce the method to upgrade detailed performance of Isochronous transfer that is one of USB transfer types. In the case that Configuration descriptor of a device has Interface descriptor that has two AlternateSetting, if Isochronous transfers are not processed smoothly due to excessive bus traffic, the application of the device changes AlternateSetting of the Interface descriptor and requires a new configuration by SetInterface() request. As a result of this adaptive configuration, the least data frame rate is guaranteed to a device that the sufficient bandwidth is not alloted. And if the bus traffic is normal, the algorithm to return to the original AlteranteSetting is introduced. this introduced method resolve the bottleneck of moving picture transfer that can occur in home network connected by multiple digital devices.

• Journal of Korean Society of Transportation
• /
• v.10 no.1
• /
• pp.55-62
• /
• 1992

교통수요는 교통정책 및 교통시설 계획의 수립 및 평가에 중요한 영향을 미치게 되므로 교통수요의 예측은 교통연구에서 중요한 부문을 차지하고 있다. 도로밑에 설치된 전자차량감지기(Electronic Vehicle Detector)로부터 자동 수집된 링크 교통량 자료(Traffic Counts)를 주요 입력자료로 이용하여 계획지역의 기종점 통행표(Origin Destination Trip Matrix)를 작성할 수 있는 기법 들이 최근 수년동안 많이 발달하게 되었다. 이러한 새로운 기법들은 가구조사(Home Inteview), 노변면접조사(Road-Side Interview)등을 토하여 조사된 자료를 기초로하는 전통적은 4단계 교통수요추정방법(Conventional 4-Stage Estimation Method)-통행발생(Generation), 통행분포(Distribution), 수단선택(Modal Split), 교통배분(Assignment)-과 비교하여 첫째로 정확도가 높은 링크 교통량 자료를 별도의 조사를 거치지 않고서도 수집이 가능하기 때문에 조사비용이 거의 들지 않아도 되어 경제적이고, 둘째로 전통적인 수요예측방법들에서 요구되어지는 복잡한 모형수립 및 계수조정(Parameter Calibration)이 필요하지 않아 간편하고 셋째로 오래전에 작성된 기종점 통행표를 단순히 링크 교통량 자료만을 이용하여 쉽게 보완할 수 있어 지속적인 자료의 축적(Data Age-ing)이 가능하며 더 나아 가서 소위 연속적인 교통 계획 및 교통시설관리(Continuous Transport Planning and Management)를 가능케 하는 등의 여러 장점 때문에 많은 주목을 받아 오고 최근 몇 년이 꾸준히 실무에 유용하게 적용이 되고 있는 실정이다. 본 연구는 링크 교통량자료를 이용하여 기종점 통행표를 작성하기 위하여 개발된 기존의 여러 기법들 가운데 특히 용량제약조건(Capacity-Restrained Condition)하에서 기존의 방법들을 상호 검토한 후 Wardrop의 교통망 평형원칙(Wardrop's First Network Equilibrium Principle)을 만족하는 새로운 추정기법을 제의하고 이의 시험결과를 논의하는 것을 주요내용으로 한다. 링크 교통량 자료를 이용하여 기종점 통행표를 작성하는 기법들의 근본 목표는 조사된 링크 교통량(Ob-served Traffic Counts)에 가장 근접한 교통망 통행 배정 링크 교통량(Assigned Link Volumes)을 재현(Re-producing)할 수 있는 기종점 통행표들 중에서 최적의 기종점 통행표를 발견하는 것이다. 따라서 교통망에서 통행자의 여행 경로 배정을 가장 잘 반영할 수 있는 현실적인(Realistic) 교통망 통행 배정 모형(Net-work Traffic Assignment Model)의 선택은 중요한 요소가 되며 특히 교통망에 교통체증(Traffic Conges-tion)이 심할 경우 교통망 통행자 평형조건(Network Traffic Equilibrium Condition)을 고려하기 위한 특별한 처리가 요구되어진다. 본 연구는 Whllumsen(Hall, Van Vliet and Willumsen, 1980)에 의하여 개발된 ME2(Maximum Entropy Matrix Estimation)기법에서 반복식 추정방법(Sequential Estimation Method)을 사용할 경우 Wardrop의 평형조건을 만족하는 기종점 통행표를 구할 수 없다는 단점을 극복하기 위한 방안으로서 엔트로피 극대화문제와 교통망 평형 조건(Entropy Maximisation and Network Equilibrium Condition)의 두 문제를 동시에 해결할 수 있는 새로운 수식모형과 이를 풀기 위한 알고리즘(Simultaneous Solution Algorithm)을 제의하였다. 제의된 수식모형과 알고리즘을 예제 교통망(Example Network)을 이용한 시험하고 그 결과를 ME2 의 반복식 추정 방법으로부터 구한 기종점 통행표와 비교 검토하였다.

• Journal of KIISE:Information Networking
• /
• v.34 no.1
• /
• pp.62-72
• /
• 2007

Mobile Virtual Private Network (MVPN) provides VPN services without geographical restriction to mobile workers using mobile devices. Coexistence of Mobile IP (MIP) protocol for mobility and IPsec-based VPN technology are necessary in order to provide continuous VPN service to mobile users. However, Problems like registration failure or frequent IPsec tunnel re-negotiation occur when IPsec-based VPN Gateway (GW) and MIP are used together. In order to solve these problems, IETF proposes a mechanism which uses external home agent (x-HA) located external to the corporate VPN GW. In addition, based on the IETF proposal, a mechanism that assigns x-HA dynamically in the networks where MN is currently located was also proposed with the purpose to reduce handover latency as well as end-to-end delay. However, this mechanism has problems such as exposure of a session key for dynamic Mobility Security Association (MSA) or a long latency in case of the handover between different networks. In this paper, we propose a new MVPN protocol in order to minimize handover latency, enhance the security in key exchange, and to reduce data losses cause by handover. Through a course of simulation, the performance of proposed protocol is compared with the existing mechanism.

• Journal of IKEEE
• /
• v.22 no.1
• /
• pp.156-161
• /
• 2018

Recently, as the population rapidly increases, food shortages and waste are emerging serious problem. In order to solve this problem, various countries and enterprises are trying research and product development such as a study of consumers' purchasing patterns of food and a development of smart refrigerator using IoT technology. However, the smart refrigerators which currently sold have high price issue and another waste due to malfunction and breakage by complicated configurations. In this paper, we proposed a low-cost smart refrigerator system based on IoT for solving the problem and efficient management of ingredients. The system recognizes and registers ingredients through QR code, image recognition, and speech recognition, and can provide various services of the smart refrigerator. In order to improve an accuracy of image recognition, we used a model using a deep learning algorithm and proved that it is possible to register ingredients accurately.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.3 s.345
• /
• pp.111-117
• /
• 2006

Developing a home network and a ubiquitous infrastructure requires various communication techniques and devices with more advanced hardware. With this development, increasing realtime access to multimedia data results in rapid degradation of qualify for multimedia playback. This paper presents a traffic allocation technique based on MPP(Media Preference for Presentation) that can steadily maintain multimedia playback quality by adaptively allocating streaming traffic requested from clients with different playback performances. Media preference is defined in accordance with content popularity and playback performance of client devices. Through experiments when requested stream data exceeds processing ability of a midea server, the proposed allocation technique shows 10% quality improvement comparing to the system without applying the proposed allocation technique.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.645-647
• /
• 2015

IoT (Internet of Things) technology has immense potential in our modern society. Generally when used in a consumer electronics environment, the IoT can continue to work more efficiently than the conventional method of home appliances. In order to use the IOT device's effectively in the environment, we need to determine the relationship between all device's based on their state and priority and compare it to find the suitable device based on the state and priority. In this paper I have done research on Wi-Fi devices in a network environment and I have observed relationships between them. Also I have proposed an algorithm to determine the device status and priority. This determines high-priority devices and they are identified on the basis of the state of the device associated with a particular operation and proceeds further.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.8B
• /
• pp.751-759
• /
• 2009

Due to the advance of WLAN technology, the use of the multimedia service such as the video streaming service has been increased in the home network. However, we need to study the method which decreases the transmission delay and the frame loss rate to provide QoS of the video streaming service. Therefore, this paper proposes an active buffer management mechanism to guarantee QoS of the streaming service in IEEE 802.11e EDCA. The proposed protocol discards the frame in the HoL of the buffer based on the importance of each frame and the virtual transmission delay of frame newly arriving at the buffer. In the simulation results, the proposed algorithm not only decreases the frame loss probability of important I and P frames but also stabilizes the transmission delay. It may increase the QoS of video streaming services.

• Journal of Satellite, Information and Communications
• /
• v.9 no.2
• /
• pp.85-90
• /
• 2014

In this paper, we propose a novel UE(User Equipment) position tracking algorithm of visible light signals, and analyze BER performance when optical beamforming scheme is applied in VLC(Visible Light Communications) system. The optical beamforming scheme as an interference mitigation method provides good optical channel conditions between UE(Rx) and LED light(Tx), and lead to better BER performance because UE is able to get a better RSSI(Received Signal strength indication) value. For the system BER, the proposed UE position tracking method for optical beamforming transmission demonstrates the performance enhancement compared to a system not applied the optical beamforming transmission scheme, and up to about 5~6dB SNR performance gain is achieved. The results of the paper can be applied to design of various VLC and LBS applications for ubiquitous home network systems.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.5
• /
• pp.1047-1057
• /
• 2024

Recently, deep neural network-based reinforcement learning models have been applied in various advanced industrial fields such as autonomous driving, smart factories, and home networks, but it has been shown to be vulnerable to malicious adversarial attack. In this paper, we applied deep reinforcement learning models, DQN and PPO, to the autonomous driving simulation environment HighwayEnv and conducted three adversarial attacks: FGSM(Fast Gradient Sign Method), BIM(Basic Iterative Method), PGD(Projected Gradient Descent) and CW(Carlini and Wagner). In order to respond to adversarial attack, we proposed a method for deep learning models based on reinforcement learning to operate normally by removing noise from adversarial images using a bilateral filter algorithm. Furthermore, we analyzed performance of adversarial attacks using two popular metrics such as average of episode duration and the average of the reward obtained by the agent. In our experiments on a model that removes noise of adversarial images using a bilateral filter, we confirmed that the performance is maintained as good as when no adversarial attack was performed.

• Journal of the Korean Geographical Society
• /
• v.44 no.6
• /
• pp.813-832
• /
• 2009

As accessibility to health-care service in less populated rural areas is geographically limited and demand for public health-care by the aging is increasing, a new approach for health-care service such as a home care service is becoming more popular. For a home care service, health-care personnels directly visit to location of health-care clients. Such changes in provision of health services require developing innovative and scientific approaches for efficient allocation of health resources and managing services by public health-care organizations. The purpose of this study is to formulate a location model for visiting health-care districting for the rural elderly and to develop an Automated Zoning Procedure (AZP) to solve this model. Mobility, workload balance and contiguity criteria are considered in the model. Three different objective functions are evaluated; 1) minimizing the sum of network distance between the unit areas in a district, 2) maximizing spatial interaction between the unit areas in a district, and 3) minimizing tour distance that visits each unit area exactly once in a district. The AZP for solving the model is developed and applied to a rural area. The application results demonstrate that the AZP can generate different districting systems for each objective functions.