• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.10 s.352
• /
• pp.8-17
• /
• 2006

Recently, the high-speed Internet users increase rapidly and broadband networks have been widely deployed. However, the current TCP congestion control algorithm was designed for relatively narrowband network environments, and thus its performance is inefficient for traffic transport in broadband networks. To remedy this problem, the TCP having an enhanced congestion control algorithm is required for broadband networks. In this paper, we propose an improved TCP congestion control that can sufficiently utilize the large available bandwidth in broadband networks. The proposed algorithm predicts the available bandwidth by using ACK information and RTT variation, and prevents large packet losses by adjusting congestion window size appropriately. Also, it can rapidly utilize the large available bandwidth by enhancing the legacy TCP algorithm in congestion avoidance phase. In order to evaluate the performance of the proposed algorithm, we use the ns-2 simulator. The simulation results show that the proposed algorithm improves not only the utilization of the available bandwidth but also RTT fairness and the fairness between contending TCP flows better than the HSTCP in high bandwidth delay product network environment.

• The KIPS Transactions:PartC
• /
• v.12C no.4 s.100
• /
• pp.559-570
• /
• 2005

ECN is accurate in determining traffic congestion since it explicitly notifies the incipient congestion. Therefore, ECN method has been thoroughly studied in the field of wireless TCP. This paper introduces a formula to find the optimal threshold for ECN marking. We have implemented a Petri net model of 'TCP with ECN strategy' and performed simulations on it in order to verify the validity of the formula. We have also introduced ideas of applying the formula in practice. The primary contribution of this paper is proposing a formula to find the optimal threshold for ECN marking. However, introducing the Petri net model of 'TCP with ECN strategy' is no less valuable contribution because it can be helpfully used by the researchers in studying network protocols. We have built the Petri net model by modifying the existing Petri net model of TCP. In order to add ECN strategy to the existing model, we have mainly modified the network part. We have also modified sender part and receiver part as well.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.6
• /
• pp.1194-1201
• /
• 2003

Measurements of network traffic have shown that self-similarity is a ubiquitous phenomenon spanning across diverse network environments. We have advance the framework of multiple time scale congestion control and show its effectiveness at enhancing performance for fast TCP prototype control. In this paper, we extend the fast TCP prototype control framework to window-based congestion control, in particular, TCP. This is performed by interfacing TCP with a large time scale control module which adjusts the aggressiveness of bandwidth consumption behavior exhibited by TCP as a function of "large time scale" network state. i.e., conformation that exceeds the horizon of the feedback loop as determined by RTT. Performance evaluation of fast TCP prototype is facilitated by a simulation bench-mark environment which is based on physical modeling of self-similar traffic. We explicate out methodology for discerning and evaluating the impact of changes in transport protocols in the protocol stack under self-similar traffic conditions. We discuss issues arising in comparative performance evaluation under heavy-tailed workload. workload.

• Journal of the Korea Society of Computer and Information
• /
• v.4 no.4
• /
• pp.148-154
• /
• 1999

TCP has improved by many papers which suggest the new algorithms and modify the previous algorithms. This paper compares Tahoe, Reno, New-Reno, Vegas, and SACK. The first version is Tahoe and is globally used. Reno has optimal performance during occurring one packet loss within a window of data, but can suffer from performance when multiple packets are dropped from a window of data. New-Reno avoids some of the performance problems of Reno TCP when multiple packets are dropped from a window of data. but is occurring the problem of the necessary retransmission. SACK resolves the all above problems and is used in bandwidth delay product environment. Vegas uses network bandwidth more efficiently and is a new implementation of TCP that achieves between 40 and 70 better throughput, with one-fifth to one-half the losses, as compared to the implementation of Reno TCP.

• Proceedings of the KIEE Conference
• /
• 2002.11c
• /
• pp.27-30
• /
• 2002

In this paper, we propose a window-based congestion control algorithm to achieve a fair sharing of the available bandwidth in ECN capable heterogeneous TCP networks with a single bottleneck link. The proposed algorithm is based on extracting the network status from the successive binary congestion information provided by ECN. From the explicit network information, we estimate the fair window size proportional to the propagation delay. Through simulations, the effect and performance of the proposed algorithm are shown for the heterogeneous networks.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10c
• /
• pp.604-606
• /
• 2004

유선망과는 달리 무선망은 폐이딩(fading), 잡음, 간섭 등에 의해 상대적으로 높은 에러율을 보인다. 이러한 무선망에서 TCP를 적용할 경우, 무선망에서의 비트 에러에 의한 패킷 손실도 혼잡으로 간주하여 송신 윈도우를 줄임으로써 비효율적인 특성을 보인다. 이러한 무선 환경에서 기존TCP의 성능을 향상시키기 위해 1-TCP, M-TCP, snoop등 다양한 기법들이 연구되고 있는데, 그 중에서 snoop은 양단간의 의미를 유지하면서도 효율적인 기법으로 널리 알려져 있다 그러나, Snoop은 무선망에서 많은 패킷이 손실되는 경우에는 비효율적인데, 본 논문에서는 snoop을 개선하여 에러율이 높은 무선망 환경에서 효과적인 성능을 발휘하는 fast Snoop 기법을 제안한다 제안된 알고리즘을 시뮬레이션을 통해 기존의 Snoop과 비교하여 성능 개선 효과를 확인 하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.319-321
• /
• 2015

인지무선네트워크 (Cognitive Radio Networks) 환경에서 스펙트럼의 원소유주인 1차사용자가 전송을 개시하는 경우, 같은 채널을 사용하는 2차사용자의 TCP (Transmission Control Protocol) 는 전송 불능 상태가 되어 심각한 성능저하가 발생한다. 이러한 성능저하는 1차사용자의 등장으로 인해 채널이 사용 불가능 해지는 상태를 패킷 손실로 판단하여 재전송 타임아웃이 발생하기 때문에 발생된다. 우리는 이 문제를 링크 또는 물리 계층 (하위계층) 과 TCP간의 크로스레이어링을 통하여 해결하고자 한다. 하위 계층은 1차사용자의 전송이 감지되면, 이를 TCP에게 시그널링하고, TCP는 이를 통해 재전송 타이머와 혼잡 윈도우를 고정시키고, 패킷 전송을 중단하도록 한다. 또, 하위계층이 가용 채널을 감지하게 되면, 재차 TCP에게 시그널링을 함으로써, 전송이 신속하게 재개되도록 한다. 제안하는 방법은 실제 USRP(Universal Software Radio Peripheral)에 구현하여 성능의 향상을 검증한다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.2
• /
• pp.191-195
• /
• 2008

This paper proposes a technique for improving IEEE 802.11e EDCA's performance by dynamically adjusting each class's contention window size based on each class's traffic amount. For providing differentiated service differently from 802.11, 802.11e EDCA maintains four classes each of which specifies different static minimum and maximum contention window sizes. Since the initial window sites significantly affect the 802.11e performance, several window adjustment schemes have been proposed. One of the schemes known as CWminAS (CWmin Adaptation Scheme) dynamically and synchronously determines the four windows' site based on the periodically measured collision rate. This method, however, can lower the send probability of high priority classes since it can't differentiate their collisions from those of low priority classes, leading to the channel underutilization. For solving this problem, we propose ACATICT(Adaptive Contention-window Adjustment Technique based on Individual Class Traffic) algorithm which adapts each class window size based on each individual collision rate rather than one average collision rate. Our simulation experiments show that ACATICT achieves better utilization by around 10% at maximum.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.4
• /
• pp.741-746
• /
• 2012

Transmission control protocol(TCP) widely used as a transport protocol in the Internet includes a loss recovery function that detects and recovers packet losses by retransmissions. The loss recovery function consists of the two algorithms; fast retransmit and fast recovery. There have been researches to avoid nonnecessary retransmission timeouts (RTOs), which leads to selective acknowledgement (SACK) option and limited transmit scheme that are standardized by IETF (Internet Engineering Task Force). Recently, a method that covers the case in which a retransmitted packet is lost again has been propsed. The method, however, is not proved in terms of the additive increase multiplicative decrease (AIMD) principle of TCP congestion control. In this paper, therefore, we analyzed the method in terms of the principle by ns-simulations.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.1
• /
• pp.15-21
• /
• 2018

Per-packet scheduling is more suitable than per-flow scheduling to reduce the flow completion time by efficiently utilizing resources in data center networks. Recently, many per-packet scheduling schemes utilizing multiple paths have been proposed. However, to mitigate the negative effect of packet reordering on TCP performance, most of the schemes require supplemental measures such as putting packets in order at the lower layer. In this study, we investigate how well DCTCP, which is a representative TCP for data center networks, performs with per-packet scheduling through simulation. Simulation results show that DCTCP keeps the queue length short but that DCTCP shows low fairness due to the way of reducing the congestion window by ECN.