• Proceedings of the Korea Information Processing Society Conference
• /
• 2004.05a
• /
• pp.1271-1274
• /
• 2004

현재 운용중인 대부분의 VPN은 모든 CPE(Customer Premise Equipment) VPN GW(Gateway)들이 Center VPN GW에 연결되어 있는 Star 구조를 취하고 있다. 이러한 구조에서는 모든 트래픽들이 항상 Center VPN GW를 거쳐서 전송되므로 비효율적인 트래픽 전송이 이루어진다. 또한 대용량의 멀티미디어 트래픽 전송이 빈번하거나 다수의 지점을 갖고 있는 기업의 경우 Center VPN GW에서의 오버헤드가 증가하게 된다. 이러한 문제를 해결하기 위한 방법으로는 IPSec의 IKE(Internet Key Exchange) 메커니즘을 이용하여 CPE VPN GW간 직접 터널을 맺어 줄 수 있으나, 터널 설립에 앞서 원격지 CPE VPN GW의 주소, 요구되는 보안 등급 등의 터널 설정에 필요한 정보를 관리자가 직접 설정해 주어야 한다. 이는 현재 DHCP와 같은 동적 IP 환경에서 운용되는 ADSL 기반의 VPN 환경에서 관리 오버헤드를 증가시키는 요인이 된다. 이에 본 논문에서는 CPE VPN GW 간 직접 터널 연결이 필요할 시에 자동적으로 제반 기능들이 수행될 수 있게 하는 주문형 터널 생성(On-demand Tunnel Creation) 메커니즘 제안한다. 시뮬레이션을 통해 제안하는 방안에 대하여 성능을 조사하였고, 이와 함께 기존의 Star VPN 구조, Full-mesh VPN 구조와 성능을 비교하였다. 시뮬레이션 결과, 제안하는 방안이 기존의 Star VPN 구조보다 확장성과 트래픽 전송효율성, Center VPN GW의 오버헤드 측면에서 우수한 성능을 보였으며 Full-mesh 구조의 VPN과 거의 비슷한 종단간 지연시간과 처리율을 보였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.453-455
• /
• 2020

본 논문은 시각 장애인을 위한 감정 음성 자막 서비스를 생성하는 종단 간(end-to-end) 감정 음성 합성 시스템(emotional text-to-speech synthesis system, TTS)의 음성 합성 속도를 높이면서도 합성음의 음질을 향상시키는 방법을 제안한다. 기존에 사용했던 전역 스타일 토큰(Global Style Token, GST)을 이용한 감정 음성 합성 방법은 다양한 감정을 표현할 수 있는 장점을 갖고 있으나, 합성음을 생성하는데 필요한 시간이 길고 학습할 데이터의 동적 영역을 효과적으로 처리하지 않으면 합성음에 클리핑(clipping) 현상이 발생하는 등 음질이 저하되는 양상을 보였다. 이를 보안하기 위해 본 논문에서는 새로운 데이터 전처리 과정을 도입하였고 기존의 보코더(vocoder)인 웨이브넷(WaveNet)을 웨이브알엔엔(WaveRNN)으로 대체하여 생성 속도와 음질 측면에서 개선됨을 보였다.

• KIPS Transactions on Computer and Communication Systems
• /
• v.7 no.2
• /
• pp.33-40
• /
• 2018

In an environment with a large network bandwidth, maximizing bandwidth utilization is an important issue to increase transmission efficiency. End-to-end transfer efficiency is significantly influenced by factors such as network, data transfer nodes, and intranet network security policies. Science DMZ is an innovative network architecture that maximizes transfer performance through optimal solution of these complex components. Among these, the data transfer node is a key factor that greatly affects the transfer performance depending on storage, network interface, operating system, and transfer application tool. However, tuning parameters constituting a data transfer node must be performed to provide high transfer efficiency. In this paper, we propose a method to enhance performance through tuning parameters of 100Gbps data transfer node. With experiment result, we confirmed that the transmission efficiency can be improved greatly in 100Gbps network environment through the tuning of Jumbo frame and CPU governor. The network performance test through Iperf showed improvement of 300% compared to the default state and NVMe SSD showed 140% performance improvement compared to hard disk.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.6
• /
• pp.336-342
• /
• 2017

Traffic can be localized by reducing the traffic load on the physical network by causing traffic to be generated at the end of the packet network. By localizing traffic, the IoT-based sensitive data-related security issues can be supported effectively. In addition, it can be applied effectively to the next-generation smart network environment without changing the existing network infrastructure. In this paper, a content priority scheme was applied to smart network-based IoT data. The IoT contents were localized to efficiently pinpoint the flow of traffic on the network to enable smart forwarding. In addition, research was conducted to determine the effective network traffic routes through content localization. Through this study, the network load was reduced. In addition, it is a network structure that can guarantee user quality. In addition, it proved that the IoT service can be accommodated effectively in a smart network-based environment.

• Journal of Digital Convergence
• /
• v.17 no.12
• /
• pp.219-228
• /
• 2019

Thanks to the development of information technology and the vitalization of smart services, the Internet of Things (IoT) technology, in which various smart devices are connected to the network, has been continuously developed. In the legacy IoT architecture, data processing has been centralized based on cloud computing, but there are concerns about a single point of failure, end-to-end transmission delay, and security. To solve these problems, it is necessary to apply decentralized blockchain technology to the IoT. However, it is hard for the IoT devices with limited computing power to mine blocks, which consumes a great amount of computing resources. To overcome this difficulty, this paper proposes an IoT architecture based on the edge computing technology that can apply blockchain technology to IoT devices, which lack computing resources. This paper also presents an operaional procedure of blockchain in the edge computing-based IoT architecture.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.3
• /
• pp.443-454
• /
• 2020

The JavaScript engine is a module that receives JavaScript code as input and processes it, among many functions that are loaded into web browsers and display web pages. Many fuzzing test studies have been conducted as vulnerabilities in JavaScript engines could threaten the system security of end-users running JavaScript through browsers. Some of them have increased fuzzing efficiency by guiding test coverage in JavaScript engines, but no coverage guided fuzzing of optimized, dynamically generated machine code was attempted. Optimized JavaScript codes are difficult to perform sufficient iterative testing through fuzzing due to the function of runtime guards to free the code in the event of exceptional control flow. To solve these problems, this paper proposes a method of performing fuzzing tests on optimized machine code by avoiding deoptimization. In addition, we propose a method to measure the coverage of runtime-guards by the dynamic binary instrumentation and to guide increment of runtime-guard coverage. In our experiment, our method has outperformed the existing method at two measures: runtime coverage and iteration by time.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.6
• /
• pp.1307-1315
• /
• 2017

While the PSK mode of the DTLS is the most efficient in terms of the performance, it is not easy to pre-distribute and manage the symmetric key pairs as the number of sensor devices increases. On the other hand, both the RPK and certificate modes offer a convenient key management tool, but they do not guarantee a good computational performance. In this paper, the end-to-end security protocol suitable for the constrained devices is proposed, based on both the ECQV certificate and the PSK mode. Namely, the initial DTLS handshake is performed using the ECQV certificate, and the subsequent DTLS handshakes with the other CoAP servers in the same group are performed using the PSK mode for the purpose of reducing the overall computational load. Furthermore, a fine-grained access control for the CoAP client can be enforced to allow access to the limited number of CoAP servers.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.2 s.302
• /
• pp.27-36
• /
• 2005

This thesis proposed an effective encryption method for the DCT-based image/video contents and made it possible to operate in a high speed by implementing it as an optimized hardware. By considering the increase in the amount of the calculation in the image/video compression, reconstruction and encryption, an partial encryption was performed, in which only the important information (DC and DPCM coefficients) were selected as the data to be encrypted. As the result, the encryption cost decreased when all the original image was encrypted. As the encryption algorithm one of the multi-mode AES, DES, or SEED can be used. The proposed encryption method was implemented in software to be experimented with TM-5 for about 1,000 test images. From the result, it was verified that to induce the original image from the encrypted one is not possible. At that situation, the decrease in compression ratio was only $1.6\%$. The hardware encryption system implemented in Verilog-HDL was synthesized to find the gate-level circuit in the SynopsysTM design compiler with the Hynix $0.25{\mu}m$ CMOS Phantom-cell library. Timing simulation was performed by Verilog-XL from CadenceTM, which resulted in the stable operation in the frequency above 100MHz. Accordingly, the proposed encryption method and the implemented hardware are expected to be effectively used as a good solution for the end-to-end security which is considered as one of the important problems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.3B
• /
• pp.408-420
• /
• 2010

On a Mobile Ad hoc NETwork(MANET), it is difficult to detect and prevent misbehaviors nodes existing between end nodes, as communication between remote nodes is made through multiple hop routes due to lack of a fixed networked structure. Therefore, to maintain MANET's performance and security, a technique to identify misbehaving middle nodes and nodes that are compromise by such nodes is required. However, previously proposed techniques assumed that nodes comprising MANET are in a friendly and cooperative relationship, and suggested only methods to identify misbehaving nodes. When these methods are applied to a larger-scale MANET, large overhead is induced. As such, this paper suggests a system model called Secure Cluster-based MANET(SecCBM) to provide secure communication between components aperANET and to ensure eed. As such, this pand managems suapemisbehavior nodes. SecCBM consists apetwo stages. The first is the preventis pstage, whereemisbehavior nodes are identified when rANET is comprised by using a cluster-based hierarchical control structure through dynamic authentication. The second is the post-preventis pstage, whereemisbehavior nodes created during the course apecommunication amongst nodes comprising the network are dh, thed by using FC and MN tables. Through this, MANET's communication safety and efficiency were improved and the proposed method was confirmed to be suitable for MANET through simulation performance evaluation.

• The Magazine of the IEIE
• /
• v.25 no.6
• /
• pp.69-78
• /
• 1998

프레임 릴레이망을 백본으로 한 LAN과 LAN접속으로 형성되는 기업통신망이 현재 활발히 전개되면서 인터넷을 통한 광고, 전자거래가 중요한 서비스로 진행되고 있다. 국내의 프레임 릴레이 서비스 현황을 살펴보면 KT, 데이콤, 일부기업등에서 프레임 릴레이 교환기를 수입하여 주고객인 기업에 서비스를 제공하고 있다. 이들은 사용자 요구에 의해 주로 LAN과 LAN에 접속된 라우터 등의 종단 시스템간 PVC(영구가상회선: Permanant Virtual Circuit)접속을 통하여 다양한 서비스를 제공하면서 독자적인 기업통신체계를 구축하고 있는 실정이다. 이와같이 사용자 요구로 PVC등록시 프레임 릴레이망과 LAN을 연결하는 링크의 QoS(서비스 질: Quality of Service), 트래픽 파라메터는 수동적으로 설정된다. 현재 국내에서는 프레임 릴레이 교환기 개발은 전무하고 단지 그들을 수입하여 운용하는 수준이며 향후 ATM(비동기 전달모드: Asynchronous Transfer Mode)망의 중요 서비스 연동이 될 프레임 릴레이 기술은 심도있게 연구되고 있다. 향후 고속통신을 제공하는 백본망은 기존의 프레임 릴레이망과 ATM망이 공존할 추세이며 점진적으로 프레임 릴레이망은 사라질 것이다. 이때 ATM 망이 기존의 프레임 릴레이망을 백본으로 한 기업통신망 수용시 원할한 서비스 제공을 위해서는 프레임 릴레이망을 백본으로 채택한 기존의 기업통신망을 ATM망에서 수용시 야기되는 문제점의 제시와 이의 해결은 매우 중요하다고 생각된다. FR/ATM(Frame Relay/ATM)연동시 기업통신망 입장에서는 PVC뿐만 아니라 SVC(교환가상회선: Switched Virtual Circuit)호처리가 수용되어야 한다. 이와같이 PVC와 SVC 수용으로 기업통신망은 통신망의 구성 및 관리가 복잡, 다양해지며, 또 그 기업은 관련된 다양하고 수많은 고객과 접속될 수 있어야 한다. 이때 ATM망을 백본으로 한 공중망이 관련 기업의 사설 정보를 관리하도록 한다는 것은 그 서비스의 이용자에게 보안문제를 제기함은 물론 서비스 이용자가 자신의 망에서 동작하는 여러 서비스를 고객 스스로가 관리하는 것이 바람직하다. CNN(고객망관리 : Customer Network Management)은 이러한 고객에 의한 통신망, 공중망에 대한 부분적 관리 및 서비스에 대한 관리가 이루어 질수 있도록 수행하는 기능이다. 본 논문에서는 ATM망이나 프레임 릴레이망과 ATM 망이 공존시 기업통신망관점에서 야기되는 문제점의 제시와 이의 해결 방안을 살펴보았다. 우선 FR/ATM 연동에서 PVC설정뿐만 아니라 SVC 설정 방법의 연구, 이때 고려된 QoS,트래픽 처리과정을 살펴보았다. 또한 망관리 입장에서 기업통신망 관리자와 ATM망 관리자와의 온-라인 서비스를 제공하기 위해서는 변화된 통신망 차원서의 확장된 Managed Object(MO)를 근간으로한 체계적인 CNN-NMS(CNM-Network management System)연동을 살펴본후 결론에서는 추후 계속연구 및 조사 되어야 할 항목을 기업통신망 입장에서 서술하여 보았다.