• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.923-926
• /
• 2000

In this paper we address the problem of detecting and estimating an unknown narrow band signal in a noise interference environment A new practical band estimation method, yielding good performance even in case of finite-length data, is presented. More specifically, wavelet packet transform is utilized to detect the more accurate time-variant band, then we estimate the power from wavelet filter-coefficients of the respective band. Also, third-order cumulants, and projection cross-correlation (PCC) criterion are utilized to achieve an effective SNR improvement for the time-variant band estimation. In case of time variant band estimation, the PCC method yields better performance than the correlation method.

• The KIPS Transactions:PartC
• /
• v.8C no.4
• /
• pp.405-414
• /
• 2001

전력연구원에서는 ATM WAN 통합망 기반 구축을 통하여 통신회선 임대비용의 절감과 안정적이고 고속의 인터넷 서비스를 수용함으로써 통신망의 경쟁력을 확보하는 것을 목표로 ATM 기반 전력통신망을 설계하였다. 이를 위하여 기존의 패킷 통신망 트래픽 통계 분석결과와 연간 트래픽 증가량 예측치를 기반으로 향후 서비스 수용계획을 고려한 토폴로지와 링크용량을 설계하였으며, 이를 바탕으로 전력통신망에서의 ATM 스위치 기반 네트워크 모델을 확립하였다. 본 논문에서는 ATM 기반 전력통신망의 설계결과를 요약한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10b
• /
• pp.1047-1050
• /
• 2000

차세대 인터넷의 요구 사항을 만족시켜 주기 위하여 IETF에서는 MPLS(Multiprotocol Label Switching) 기술을 표준화하고 있다. MPLS는 ATM과 흡사한 레이블을 사용하여 레이블 스왑핑에 의한 포워딩을 수행하므로 IP 패킷을 매우 빠른 속도로 전달해 줄뿐만 아니라 ATM 망에서의 다양한 질의 서비스 제공을 가능하게 하여 기존의 ATM 망의 활용도를 증가시킨다. 본 논문에서는 현재 개발 진행 중인 ATM 스위치 기반 MPLS LER 시스템에서 성능 관리 기능을 위한 구조를 제시하고, 구현 내용을 기술한다. 본 시스템에서 제공되는 통계 대상에는 MPLS 연결인 LSP 트래픽과 Non_ MPLS 도메인의 연결인 가입자 트래픽 및 제어 트래픽 등이 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07a
• /
• pp.142-144
• /
• 2005

Distributed Denial of Service는 네트워크나 개인 호스트를 위협하는 대표적인 공격 트래픽이다. DDoS 공격은 특정한 패턴을 가지고 있지 않기 때문에 탐지가 어려울 뿐 아니라, TNF2K와 같은 간단한 tool로 공격이 가능하여 그 심각성은 실로 크다. 이러한 DDoS를 탐지하기 위한 메카니즘이나 알고리즘은 많이 개발되었다. 하지만 DDoS의 근원지를 판별하고 대응하는 것이 아닌, 단지 방어 지정에서 전체 threshold를 낮추거나 leaky bucket처럼 수용 능력 이상의 패킷을 폐기하는 방법으로 네트워크나 개인 호스트를 보호한다. 무분별하게 전체 트래픽을 줄이는 것은 네트워크의 resource를 고갈 시키지는 않지만, 정상적인 clients가 공격당하고 있는 호스트에 연결을 할 수가 없다. 이를 위해 여러 단계의 테스트를 통해 합법적인 검증 IP table을 만들고, 검증 IP table에 있는 source IP를 제외한 나머지 트래픽을 차단한다면, DDoS 공격에 대해서 대응을 하면서 정상적인 Clients의 연결을 보호 할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.133-134
• /
• 2009

USN은 저전력 무선 네트워크와 마이크로 센서를 결합하여 환경이나 상황을 인지하고 모니터링을 통해 수집된 정보를 사람에게 전달 하는 기술로, 연구를 위해 다양한 시뮬레이터가 개발되었다. 그러나 기존의 시뮬레이터는 패킷 검증을 위해 설계되고 개발되어 센서의 탐지 동작이 터미널 노드에서 임의로 발생 시키는 한계점을 가지고 있었다. 본 논문에서 제안하는 CROWD는 이동 객체 기법을 이용하여, 센서의 동적 동작을 보장하고, 노드의 수명 예측, 에너지 소모 효율의 통계, 노드 배치의 변화에 따른 에너지 소모량 변화를 측정한다.

• The KIPS Transactions:PartC
• /
• v.18C no.4
• /
• pp.243-250
• /
• 2011

Nowadays, the traffic type and behavior are extremely diverse due to the appearance of various services on Internet, which makes the need of traffic identification important for efficient operation and management of network. In recent years traffic identification methodology using statistical features of flow has been broadly studied. We also proposed a traffic identification methodology using payload size distribution in our previous work, which has a problem of low completeness. In this paper, we improved the completeness by solving the PSD conflict using IP and port. And we improved the accuracy by changing the distance measurement between flow and statistic signature from vector distance to per-packet distance. The feasibility of our methodology was proved via experimental evaluation on our campus network.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.40 no.12
• /
• pp.9-17
• /
• 2003

In this paper we proposea method for estimating the bandwidth in next-generation If network. Especially, we concentrate on the edge routers accommodating the VoIP connections as well as a group of data connections. Bandwidth dimensioning is carried out at call level and packet level for voice traffic in the next-generation IP network. The model incorporates the statistical estimation approach at a call level for obtaining the number of voice connections simultaneously in the active mode. The call level model incorporates a statistical technique to compute the statistics of the number of active connections such as the mean and variance of the simultaneously connected calls in the network. The packet level model represents a load map for voice and data traffic by using non-preemptive M/G/1 queuing model with strict priority for voice over data buffer, From the proposed traffic model, we can derive a graph for upper bounds on the traffic load in terms of bandwidth for voice and data connections. Via numerical experiments we illustrate the implication of the work.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.108-110
• /
• 2021

Since 2015, attacks using the IoT protocol have been continuously reported. Among various IoT protocols, attackers attempt DDoS attacks using SSDP(Simple Service Discovery Protocol), and as statistics of cyber shelters, Korea has about 1 million open SSDP servers. Vulnerable SSDP servers connected to the Internet can generate more than 50Gb of traffic and the risk of attack increases gradually. Until recently, distributed denial of service attacks and distributed reflective denial of service attacks have been a security issue. Accordingly, the purpose of this study is to analyze the request packet of the existing SSDP protocol to identify an amplification attack and to avoid a response when an amplification attack is suspected, thereby preventing network load due to the occurrence of a large number of response packets due to the role of traffic reflection amplification.

• Journal of Broadcast Engineering
• /
• v.14 no.5
• /
• pp.573-588
• /
• 2009

In this paper, we propose a novel video quality degradation monitoring scheme titled VR-VQMS(Visual Rhythm based Video Quality Monitoring Scheme) over an IPTV service prone to packet losses during network transmission. Proposed scheme quantifies the amount of quality degradation due to packet losses, and can be classified into a RR(reduced-reference) based quality measurement scheme exploiting visual rhythm data of H.264-encoded video frames at a media server and reconstructed ones at an Set-top Box as feature information. Two scenarios, On-line and Off-line VR-VQMS, are proposed as the practical solutions. We define the NPSNR(Networked Peak-to-peak Signal-to-Noise Ratio) modified by the well-known PSNR as a new objective quality metric, and several additional objective and subjective metrics based on it to obtain the statistics on timing, duration, occurrence, and amount of quality degradation. Simulation results show that the proposed method closely approximates the results from 2D video frames and gives good estimation of subjective quality(i.e.,MOS(mean opinion score)) performed by 10 test observers. We expect that the proposed scheme can play a role as a practical solution to monitor the video quality experienced by individual customers in a commercial IPTV service, and be implemented as a small and light agent program running on a resource-limited set-top box.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.1
• /
• pp.23-34
• /
• 2022

With the development of network technology, the application area has also been diversified, and protocols for various purposes have been developed and the amount of traffic has exploded. Therefore, it is difficult for the network administrator to meet the stability and security standards of the network with the existing traditional switching and routing methods. Software Defined Networking (SDN) is a new networking paradigm proposed to solve this problem. SDN enables efficient network management by programming network operations. This has the advantage that network administrators can flexibly respond to various types of attacks. In this paper, we design a threat level management module, an attack detection module, a packet statistics module, and a flow rule generator that collects attack information through the controller and switch, which are components of SDN, and detects attacks based on these attributes of SDN. It proposes a method to block denial of service attacks (DoS) of advanced attackers by programming and applying honeypot. In the proposed system, the attack packet can be quickly delivered to the honeypot according to the modifiable flow rule, and the honeypot that received the attack packets analyzed the intelligent attack pattern based on this. According to the analysis results, the attack detection module and the threat level management module are adjusted to respond to intelligent attacks. The performance and feasibility of the proposed system was shown by actually implementing the proposed system, performing intelligent attacks with various attack patterns and attack levels, and checking the attack detection rate compared to the existing system.