• Journal of KIISE:Information Networking
• /
• v.32 no.5
• /
• pp.583-592
• /
• 2005

An important factor influencing the robustness of the Internet is the end-to-end TCP congestion control. However, the congestion control scheme of TCP Reno, the most popular TCP version on the Internet, employs passive congestion indication. It makes the network congestion worse. Brakmo and Peterson proposed a congestion control algorithm, TCP Vegas, by modifying the congestion avoidance scheme of TCP Reno. Many studies indicate that Vegas is able to achieve better throughput and higher stability than Reno. But there are three unfairness problems in Vegas. These problems hinder the spread of Vegas in the current Internet. In this paper, in order to solve these unfairness problems, we propose a new congestion control algorithm called TCP NewVegas. The proposed NewVegas is able to solve these unfairness problems effectively by using the variation of the number of queued packets in a bottleneck router. To evaluate the proposed approach, we compare the performance among NewVegas, Reno and Vegas. Through the simulation, NewVegas is shown to be able to achieve throughput and better fairness than Vegas.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.11
• /
• pp.3097-3107
• /
• 1999

Recently many studies have been done on the Random Early Detection(RED) algorithm as an active queue management and congestion avoidance scheme in the Internet. In this paper we first overview the characteristics of RED and the modified RED algorithms in order to understand the current status of these studies. Then we analyze the RED dynamics by investigating how RED parameters affect router queue behavior. We show the cases when RED fails since it cannot react to queue state changes aggressively due to the deterministic use of its parameters. Based on the RED parameter analysis, we propose a self-adaptive algorithm to cope with this RED weakness. In this algorithm we make two parameters be adjusted themselves depending on the queue states. One parameter is the maximum probability to drop or mark the packet at the congestion state. This parameter can be adjusted to react the long burst of traffic, consequently reducing the congestion disaster. The other parameter is the queue weight which is also adjusted aggressively in order for the average queue size to catch up with the current queue size when the queue moves from the congestion state to the stable state.

• Journal of the Korean Society for Marine Environment & Energy
• /
• v.1 no.1
• /
• pp.47-55
• /
• 1998

In this paper, authors try to assess the marine traffic environment from the viewpoint of the degree of congestion. First, the marine traffic flow simulation is carried out to get the information related to the assessment. The simulation is performed based on the wide investigation of marine traffic environment and some statistical methods. Then the degree of congestion is reasoned by using the reasoning methods of the degree of collision risk made to develop collision avoidance system. The quantified degree of congestion reasoned by this method provides much information needed to the design and the evaluation of marine traffic environment.

• Journal of Information Processing Systems
• /
• v.7 no.3
• /
• pp.531-542
• /
• 2011

In this paper, we propose a new congestion control scheme for high-speed networks. The basic idea of our proposed scheme is to adopt a game theory called, "Minority Game" (MG), to realize a selective reduction of the transmission speed of senders. More concretely, upon detecting any congestion, the scheme starts a game among all senders who are participating in the communication. The losers of the game reduce the transmission speed by a multiplicative factor. MG is a game that has recently attracted considerable attention, and it is known to have a remarkable property so that the number of winners converges to a half the number of players in spite of the selfish behavior of the players to increase its own profit. By using this property of MG, we can realize a fair reduction of the transmission speed, which is more efficient than the previous schemes in which all senders uniformly reduce their transmission speed. The effect of the proposed scheme is evaluated by simulation. The result of simulations indicates that the proposed scheme certainly realizes a selective reduction of the transmission speed. It is sufficiently fair compared to other simple randomized schemes and is sufficiently efficient compared to other conventional schemes.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.1
• /
• pp.39-43
• /
• 2007

In high speed communication networks, the determination of a transmission rate is critical for the stability of a closed-loop network system with the congestion control scheme. In ATM networks, the available bit rate (ABR) service is based on a feedback mechanism, i.e., the network status is transferred to the ABR source by a resource management (RM) cell. RM cells contain the traffic information of the downstream nodes for the traffic rate control. However, the traffic status of the downstream nodes can not be directly transferred to the source node in the IP based networks. In this paper, a new rate-based congestion control scheme using an exponential weighted moving average algorithm is proposed to build an efficient feedback control law for congestion avoidance in high speed communication networks. The proposed congestion control scheme assures the stability of switch buffers and higher link utilization of the network. Moreover, we note that the proposed congestion scheme can flexibly work along with the increasing number of input sources in the network, which results in an improved scalability.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.4
• /
• pp.2002-2019
• /
• 2017

RPL routing protocol for low-power and lossy networks is an Internet Engineering Task Force (IETF) recommended IPv6 based protocol for routing over Low power Lossy Networks (LLNs). RPL is proposed for networks with characteristics like small packet size, low bandwidth, low data rate, lossy wireless links and low power. RPL is a proactive routing protocol that creates a Directed Acyclic Graph (DAG) of the network topology. RPL is increasingly used for Internet of Things (IoT) which comprises of heterogeneous networks and applications. RPL proposes a single path routing strategy. The forwarding technique of RPL does not support multiple paths between source and destination. Multipath routing is an important strategy used in both sensor and ad-hoc network for performance enhancement. Multipath routing is also used to achieve multi-fold objectives including higher reliability, increase in throughput, fault tolerance, congestion mitigation and hole avoidance. In this paper, M-RPL (Multi-path extension of RPL) is proposed, which aims to provide temporary multiple paths during congestion over a single routing path. Congestion is primarily detected using buffer size and packet delivery ratio at forwarding nodes. Congestion is mitigated by creating partially disjoint multiple paths and by avoiding forwarding of packets through the congested node. Detailed simulation analysis of M-RPL against RPL in both grid and random topologies shows that M-RPL successfully mitigates congestion and it enhances overall network throughput.

• The KIPS Transactions:PartC
• /
• v.13C no.3 s.106
• /
• pp.353-358
• /
• 2006

Unlike TCP Reno, TCP Vegas recognizes network congestion through the measuring of RTT (Round Trip Time) and decides the main congestion control parameters, such as Windows size. But, congestion avoidance scheme of Vegas poorly reflects asymmetric characteristics of packet path because TCP Vegas uses the measuring of RTT that reflects forward/backward packet transmission delay as a forward delay. The RTT can't infer the forward/backward transmission delay variation because it only measures the packet's turn around time. In this paper, We have designed and implemented a new Vegas congestion control algorithm that can distinguish forward/backward network congestion. We have modified the source codes of TCP Vegas in Linux 2.6 kernel and verified their performance.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.474-476
• /
• 1998

전송 프로토콜은 응용 프로그램과 네트워크의 인터페이스로서, 응용 프로그램에서 요구한 QoS(Quality of Service)를 제공하는 역할을 한다. 이 중 TCP는 인터넷의 전송 흐름 제어를 위해서 사용되는 프로토콜이다. TCP의 흐름 제어를 위해서 수신된 데이터의 ACK(acknowledgement)에 따라 허가된 윈도크기만큼의 데이터를 보내는 크레디트 할당 밀집 윈도(congestion window)를 사용한 slow-start 알고리즘을 사용하며, 손실된 데이터를 재전송하기 위한 방법으로 빠른 재전송 및 회복 알고리즘을 사용한다. 본 논문에서는 빠른 재전송 알고리즘에서 나타나는 문제점을 알아보고, 이 알고리즘이 빠른 시간에 데이터 손실을 회복하고 데이터를 보낼 수 있도록 수정한 알고리즘을 소개한다. 또한 수정된 알고리즘을 확장하여 네트워크의 상태에 따라 더 많은 데이터를 보낼 수 있도록 개선한 알고리즘을 제안한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.374-376
• /
• 1998

ABR서비스는 ATM 망의 이용률을 증대시키고, 매우 버스트한 트래픽 응용을 지원하기위해 설계되었다. ATM 포럼에서는 망에서 소스의 셀 전송률을 조절하고 셀 손실을 최소화하기 위한 ABR 서비스 표준으로 폐쇄루프와 전송률기반의 폭주 제어 알고리즘을 채택하였다. 이러한 폭주 제어 알고리즘 중에서 대표적인 것이 ERIVCA(Explicit Rare Indication for Congestion Avoidance) 이다. 본 논문에서는 ERIVCA 의 견고성, 공평성 LAN/WAN 환경에 대해서 성능을 시험하고. 그 결과로 성능을 향상시키기 위해 개선된 최대-최소 구조를 제공한다. 또한 제안된 구조에 대한 시뮬레이션을 하여 기존 ERIVCA 알고리즘의 성능과 비교 분석한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.273-275
• /
• 1998

최근 무선통신의 대중화가 증가함에 따라 이동 컴퓨팅 환경에 대한 관심이 나날이높아가고 있지만 무선망의 특징을 고려한 프로토콜의 개발은 미흡한 상태이다. 유선망에서 널리 통용되고 있는 TCP는 packet 손실의 원인을 혼잡(congestion)에 의한 것으로 간주하여 혼합제어(congestion control)나 회피(avoidance)메카니즘을 이용하여 packet 의 회복 (recovery)에 적절히 대응하고 있다. 하지만 무슨망에서의 packet 손실은 handoff나 높은 bit 에러율에 의한 것으므로 기존의 TCP를 그대로 적용하게 되면 불필요한 메카니즘 호출로 인한 throughput의 저하를 가져오게 된다. 이런 성능 저하를 해결하기 위하여 무선망에 적합하도록 TCP를 개선하는 움직임이 세계적으로 일어나고 있으며, 이를 크게 단대한 프로토콜, split-connection 프로토콜, 링크 계층 프로토콜로 나눌 수 있다. 본 논문은 이러한 흐름의 여러 가지 대표적인 프로토콜들을 비교.분석하였다.