• Journal of Internet Computing and Services
• /
• v.12 no.2
• /
• pp.123-133
• /
• 2011

This paper introduces security problems on the video surveillance systems where the network cameras are equipped at remote places and isolated from the updated and secure environment and proposes a framework for a proxy server that is delegated to connect to network cameras by providing a secure connections from the clients. The server in the framework is deployed within a secure network, secretes the information for connection to cameras and authenticates the clients. Additionally, it provides a secure video service incorporating multi-level privileges for both images and clients through a encryption key distribution and management facility. Through an implementation of the server and a its deployment, it was proved that In this server implement to multi network camera and we confirm compare direct access to network camera equal video quality of service and it can be protection network camera. We expect that can be secure and integral management about traditional network camera through experimental result.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11a
• /
• pp.361-363
• /
• 2005

최근의 컴퓨터 네트워킹 환경에서 무선 멀티미디어 응용 서비스에 대한 사용자의 요구는 날로 증가하고 있다. 그러므로 사용자에게 원활한 서비스를 제공위해서 QoS 보장을 위한 무선 네트워크 프로토콜의 설계는 중요하다. 본 논문에서는 무선 인터넷에서 QoS를 보장하기 위한 표준인 IEEE 802.11e[1]의 문제점들을 분석하고, Multi-Hop Ad Hoc Network에서 성능 향상을 위하여 Single-channel Jamming 신호를 기반으로 한 SJMAC(Sing1e-channel Jamming MAC) 프로토콜을 설계하고, NS2 모듈을 구현한 후에 성능 평가를 하였다. 기존 IEEE 802.11e과의 성능 비교를 위하여 NS2 (Network Simulator2)[2]에 SJMAC 모듈을 추가하여 제안된 프로토콜의 성능을 검증한다.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.29A no.11
• /
• pp.70-77
• /
• 1992

Generally, the conventional tracking algorithms are very limited in the practical applications because of that the computation load is exponentially increased as the number of targets being tracked is increase. Recently, to overcome this kind of limitation, some new tracking methods based on neural network algorithms which have learning and parallel processing capabilities are introduced. By application of neural networks to multi-target tracking problems, the tracking system can be made computationally independent of the number of objects being tracked, through their characteristics of massive parallelism and dense interconnectivity. In this paper, a new neural network tracking algorithm, which has capability of adaptive target tracking with little increase of the amount of calculation under the clutter and noisy environments, is suggested and the possibility of real-time multi-target tracking system based on neural networks is also demonstrated through some good computer simulation results.

• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.56-59
• /
• 2017

CNN(Convolutional Neural Network)을 이용하여 발화 주제 다중 분류 task를 multi-labeling 방법과, cluster 방법을 이용하여 수행하고, 각 방법론에 MSE(Mean Square Error), softmax cross-entropy, sigmoid cross-entropy를 적용하여 성능을 평가하였다. Network는 음절 단위로 tokenize하고, 품사정보를 각 token의 추가한 sequence와, Naver DB를 통하여 얻은 named entity 정보를 입력으로 사용한다. 실험결과 cluster 방법으로 문제를 변형하고, sigmoid를 output layer의 activation function으로 사용하고 cross entropy cost function을 이용하여 network를 학습시켰을 때 F1 0.9873으로 가장 좋은 성능을 보였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.11
• /
• pp.2432-2438
• /
• 1997

This paper proposes recognition system using multi-backpropagation neural networks rather than single backpropagation neural network to enhance the rate of character recognition resultsing from extracting the region of velhicle number in that the image of vehicle number plate from CCD camera has a distinguish feature, that is, illumination of a pattern. The experiment in this paper shows an output that the method using multi-backpropagation neural networks rather than signal backpropagation neural network takes less training time for computation and also has higher recognition rage of vehicle number.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.3C
• /
• pp.251-258
• /
• 2012

In this paper, we propose a novel method of multi-node frequency synchronization for distributed networks. The proposed method synchronizes carrier frequencies of all nodes in the network and this enables new entry node to synchronize immediately. Moreover, when several groups exist in the network, inter-group synchronization method is proposed. The proposed distributed frequency synchronization method is expected to be very useful for the military operation scenario that new node entry is in a state of flux and group merging and splitting frequently happen.

• Proceedings of the KIEE Conference
• /
• 2005.07d
• /
• pp.2986-2988
• /
• 2005

To improve the reliability of power-line communication systems, the measurement and analysis has been proceeded in many power-line channel environments. In spite of the wired channel, power line channel has many multi-paths that are changing with load-variation, line-interconnection, impedance mismatching and so on. We accordingly need an analysis method based on the multi-path channel impulse response. Recently, a method to describe the homogeneous Power-line network has been published[1]. In this paper the modified method that can describe both the homogeneous and non -homogeneous power-line network has presented.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.7 no.5
• /
• pp.209-217
• /
• 2012

In the present, Wireless Sensor Network(WSN) has been used for the purpose of the military operation with surveillance systems and for collecting useful information from the natural environment. Basically, low-power, easy deployment and low cost are the most important factors to be deployed for WSNs. Lots of researches have been studied to meet those requirements, especially on the node capacity and battery lifetime improvements. Recently, the study of wireless mesh networks applied into the surveillance systems has been proceeded as a solution of easy deployment. In this paper, we proposed large-scale intelligent multi-layer surveillance systems based on QoS assuring Wireless Mesh Networks and implemented them in the real testbed environment. The proposed system explains functions and operations for each subsystem as well as S/W and H/W architectures. Experimental results are shown for the implemented subsystems and the performance is satisfactory for the surveillance system. We can identify the possibility of the implemented multi-layer surveillance system to be used in practice.

• IMMUNE NETWORK
• /
• v.14 no.2
• /
• pp.73-80
• /
• 2014

Because autoimmune diseases (AIDs) result from a complex combination of genetic and epigenetic factors, as well as an altered immune response to endogenous or exogenous antigens, systems biology approaches have been widely applied. The use of multi-omics approaches, including blood transcriptomics, genomics, epigenetics, proteomics, and metabolomics, not only allow for the discovery of a number of biomarkers but also will provide new directions for further translational AIDs applications. Systems biology approaches rely on high-throughput techniques with data analysis platforms that leverage the assessment of genes, proteins, metabolites, and network analysis of complex biologic or pathways implicated in specific AID conditions. To facilitate the discovery of validated and qualified biomarkers, better-coordinated multi-omics approaches and standardized translational research, in combination with the skills of biologists, clinicians, engineers, and bioinformaticians, are required.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2003.05a
• /
• pp.262-269
• /
• 2003

Recently, a multi facility, multi product and multi period industrial problem has been widely investigated in Supply Chain Management (SCM). One of the key issues in the current SCM research area involved reducing both production and distribution costs. The purpose of this study is to determine the optimum quantity of production and transportation with minimum cost in the supply chain network. We have presented a mathematical model that deals with real world factors and constructs. Considering the complexity of solving such model, we have applied the genetic algorithm approach for solving this model computational experiments using a commercial genetic algorithm based optimizer. The results show that the real size problems we encountered can be solved In reasonable time