• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11a
• /
• pp.433-435
• /
• 2005

본 연구에서는 차세대 인터넷 환경의 주요 특징 중 하나인 이기종 네트워크가 혼재되어 있는 상황하에서 핸드오프가 일어나는 경우 여러 TCP 버전들(Newreno, Sack, Westwood, Vegas)이 겪게 되는 성능 변화를 분석하였다. 큰 대역폭의 네트워크와 작은 대역폭의 네트워크를 구성하여 실험해 본 결과 하부 계층의 핸드오프 지연이 없는 이상적인 상황에서도 대부분의 TCP들이 성능 저하를 겪었다. TCP 버전들 중 Vegas가 대역폭이 커지는 경우와 작아지는 경우 모두에서 가장 좋은 성능을 보였다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.2
• /
• pp.28-37
• /
• 2017

Multipath TCP (MPTCP), which is a layer 4 protocol that can get the reliability and the efficiency of the transmission by using multipath transfer, was standardized by the IETF. MPTCP provides superior performance when compared to a single TCP when used in a homogeneous network with similar network characteristics. However, MPTCP degrades performance when used in heterogeneous networks with different network characteristics. In this paper, we propose 'Delay-alerted path-blocking scheduler'. It measures the delay of each path and blocks the path with a long delay to reduce the order of packets in the receive buffer. If the duplicated packet is sent to the blocked path to measure the delay and the congestion on the blocking path is reduced, the blocked path is unblocked. For performance analysis, the proposed scheduler was implemented in the Linux kernel and improved performance was obtained in the test bed. We also confirmed that the proposed scheduler reduces the degradation of MPTCP performance in real wireless networks with heterogeneous path characteristics.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.7B
• /
• pp.711-721
• /
• 2004

To provide the end-to-end service differentiation for assured services, the random early demotion and promotion (REDP) marker in the edge router at each domain boundary monitors the aggregate flow of the incoming in-profile packets and demotes in-profile packets or promotes the previously demoted in-profile packets at the aggregate flow level according to the negotiated interdomain service level agreement (SLA). The REDP marker achieves UDP fairness in demoting and promoting packets through random and early marking decisions on packets. But, TCP fairness of the REDP marker is not obvious as fur UDP sources. In this paper, to improve TCP fairness of the REDP marker, we combine a dropper with the REDP marker. To make packet transmission rates of TCP flows more fair, at the aggregate flow level the combined dropper drops incoming excessive in-profile packets randomly with a constant probability when the token level in the leaky bucket stays In demotion region without incoming demoted in-profile packets. It performs a dropping in the demotion at a domain boundary only if there is no prior demotion. The concatenate dropping at multiple domains is avoided to manifest the effect of a dropping at a domain boundary on TCP fairness. We experiment with the REDP marker with the combined dropper using ns2 simulator for TCP sources. The simulation results show that the REDP marker with the combined dropper improves TCP fairness in demoting and promoting packets by generating fair demoted in-profile traffic compared to the REDP marker. The effectiveness of the selected drop probability is also investigated with showing its impact on the performance of the REDP marker with the combined dropper.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.6B
• /
• pp.411-419
• /
• 2008

As the set of new mobile data networking service technologies, such as WiBro, mobile WiMAX and WCDMA-HSDPA, have been introduced in recent years, IP mobility management protocols which aim to provide seamless IP service over the mobile networking environment become one of the important building blocks for the successful mobile networks. IETF has introduced a basic mobility management protocol over IPv6 (MIPv6) and a set of enhanced protocols such as FMIPv6 and HMIPv6, followed by FHMIPv6, S-MIPv6 and HIMIPv6 which combine the benefits of the basic protocols. We have studied the impact of the various IP mobility management protocols over TCP sessions through extensive simulations with ns-2. Our simulation results show that TCP performance with the HIMIPv6 outperforms MIPv6, FMIPv6 and HMIPv6 over the various TCP protocols, due to its efficient combination of the hierarchical mobility management scheme of the HMIPv6, the predictive handover scheme of the FMIPv6 and the fast simulcasting scheme of the S-MIP.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.6A
• /
• pp.667-667
• /
• 2004

We must design the buffer algorithm that protects traffic congestion and decreasing throughput at satellite communication network. It is important that buffer algorithm is satisfied with the good performance of transmission packet, responsibility of many connecting traffic and the QOS for connecting character. Old buffer algorithms are not the suitable algorithms when we have the satellite communication network environment. RED buffer algorithm is proposed by Floyd. It has a better performance than old buffer algorithm. But this algorithm is not well adapted a number of connecting TCP packet and changing network, so this algorithm has a bad performance on satellite communication network that is many of connecting user at same time. This paper propose the TARED(Target Adaptive RED). It has a good performance, adaptation and stability on satellite communication network and has not overflow and underflow of the buffer level.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.6A
• /
• pp.666-676
• /
• 2004

We must design the buffer algorithm that protects traffic congestion and decreasing throughput at satellite communication network. It is important that buffer algorithm is satisfied with the good performance of transmission packet, responsibility of many connecting traffic and the 005 for connecting character. Old buffer algorithms are not the suitable algorithms when we have the satellite communication network environment. RED buffer algorithm is proposed by Floyd. It has a better performance than old buffer algorithm. But this algorithm is not well adapted a number of connecting TCP packet and changing network, so this algorithm has a bad Performance on satellite communication network that is many of connecting user at same time. This paper Propose the TARED(Target Adaptive RED). It has a good performance, adaptation and stability on satellite communication network and has not overflow and underflow of the buffer level.

• New & Renewable Energy
• /
• v.7 no.3
• /
• pp.92-100
• /
• 2011

With the increased demand of clean energy and global warming measures, the renewable energy development has been increased recently. The TCP (Tidal Current Power) is one of the ocean renewable energy sources. Having the high tidal energy source in Korea, there are many potential TCP sites with strong current speed. The rotor, which initially converts the energy, is a very important component because it affects the efficiency of the entire system. The rotor performance is determined by various design parameters including number of blades, shape, sectional size, diameters and etc. However, the interactions between devices also contribute significantly to the energy production. The rotor performance considering the interaction needs to be investigated to predict the exact power in the farm. This paper introduces the optimum design of TCP turbine and the performance of devices considering the interference between rotors.

• The Journal of the Korea Contents Association
• /
• v.10 no.3
• /
• pp.44-52
• /
• 2010

The most popular world wide web traffic in the Internet uses TCP as the transport layer protocol. Since TCP utilizes the single path, it can not communicate with the correspondent host during the link-down. On the other hand, SCTP can still communicate with the other SCTP entity by using alternate path even while the primary path is down. Most of previous studies have conducted the performance comparison research between TCP and SCTP by using typical file transfer. Since web traffic with self-similarity is characterized by the packet inter-arrival times and shape parameter affecting the size of web file in the Pareto distribution, it is necessary to perform the experiments considering these parameters. This paper aims to compare the throughput between TCP and SCTP while varying parameters reflecting the web traffic characteristics in link-down environment. Experimental results for web traffic using NS-2 simulator show that the throughput of SCTP using multi-homing is better than that of TCP. Simulation also shows that TCP is more affected than SCTP by mean inter-arrival and shape parameters with regard to the web traffic. These results can be applied to estimate the performance variation of web traffic due to the duration of link-down.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.05a
• /
• pp.252-255
• /
• 2002

As more critical services are provided in the internet, the risk to these services from malicious users increases. Several networks have experienced problems like Denial of Service(DoS) attacks recently. We analyse a network-based denial of service attack, which is called SYM flooding, to TCP-based networks. It occurs by an attacker who sends TCP connection requests with spoofed source address to a target system. Each request causes the targeted system to send instantly data packets out of a limited pool of resources. Then the target system's resources are exhausted and incoming TCP port connections can not be established. The paper is concerned with a detailed analysis of TCP SYN flooding denial of service attack. In this paper, we propose a Real Time Scan Detector(RTSD) mechanism and evaluate it\`s Performance.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.7B
• /
• pp.431-437
• /
• 2007

Web proxy caching provides an effective way to reduce access latency and bandwidth requirement. In particular, prefix caching is considered as an alternative for improving video delivery over wide area networks because video objects are usually too large to be cached in their entirety. Nevertheless, many studies have pointed that the user-perceived latency is often not dominated by object transmission time, but rather by setup process such as TCP connection time that precedes it. We propose pre-connecting techniques and show that the techniques can be used efficiently in TCP splicing. Our analysis shows the pre-connection significantly reduces start-up latency and TCP connection time in simple analytical model. The deployment of the proposed pre-connection does not require protocol modification or the cooperation of other entities.