• Journal of KIISE:Information Networking
• /
• v.31 no.1
• /
• pp.20-26
• /
• 2004

A congestion control algorithm to prevent buffer overflow in MSR(Mobility Support Router) for I-TCP is proposed. Due to high bit error rate and frequent hand-offs over wireless environment, the current congestion control scheme in TCP Reno over mixed(wired and wireless) network exhibits lower throughput than the throughput achieved over wired only network. I-TCP has been proposed to address this by splitting a TCP connection into two TCP connections over wired section and wireless section, respectively. However, buffer overflow in MSR may occur whenever there are excessive bit errors or frequent hand-offs. This may lead to the loss of packets acked by MSR(resident in buffer) to the sender, but not received by the receiver, breaking TCP end-to-end semantics. In this Paper, a new scheme is proposed to prevent the MSR buffer from overflow by introducing “flow control” between the sender and the MSR. Advertised window for the TCP connection between the sender and the MSR is tied to the remaining MSR buffer space, controlling the flow of packets to the MSR buffer before overflow occurs.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.41 no.4
• /
• pp.35-44
• /
• 2004

Recently announced TCP Vegas predicts the degree of congestion in the network and then control the congestion window size. Thus it shows better performance than TCP Reno. however, TCP vegas does not assume any network model, its congestion window control is very limited. Because or this limitation, TCP vegas still can not adapt to fast changing available bandwidth. In this paper, we introduce a new TCP algorithm which adapts to fast changing available bandwidth well. To devise such a TCP, we model the end to end network of TCP connection as a queueing system and finds congestion window size which can utilize the available bandwidth sufficiently but not make the network congested. The simulation results show that our algorithm adapts to the avaliable bandwidth faster than TCP vegas and as a results, when the available bandwidth is changing rapidly, our algorithm not only operates more stably than TCP Vegas, but also it shows higher thruput than TCP Vegas.

• 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 Korea Information Processing Society Conference
• /
• 2003.11b
• /
• pp.1145-1148
• /
• 2003

최근의 인터넷에서의 데이터 흐름을 보면 비대칭흐름의 경향이 있다. 비대칭흐름은 주로 하향링크의 데이터 흐름이 많은 것이 특징이며 하향링크에서 데이터흐름을 저해하는 요소인 응답(acknowledgement)을 줄여나가는 기법이 필요하다. 본 논문에서는 순방향에서의 TCP 패킷 처리율과 역방향에서의 ACK 패킷 처리율에 초점을 맞추어 수신된 패킷 순서번호를 TCP 버전별로 관찰해 보고, 실제로 처리된 패킷의 처리율로 비교 평가하였다. 시뮬레이션 결과로, 비대칭링크 유선망에서 Tahoe 버전이 Sack, NewReno보다 성능이 좋다는 것을 알 수 있었다.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.88-91
• /
• 2000

In this paper we approach the problem of congestion control for TCP traffic over ATM-UBR networks by focusing on the fact that to get best performance. We study how to efficiently support TCP traffic in the subnet ATM model, when ATM is only a single link in the whole path. We show that when UBR connection. We analyze the ATM-UBR network service using the BSD 4.3 Reno, Tahoe TCP. However we found the fact that the characteristic of fast recovery algorithm makes a serious degradation of performance in multiple cell loss drop situation. We propose new fast recovery algorithm to solve the problem.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.3
• /
• pp.317-323
• /
• 2007

Quality of Service (QoS) in the Internet depends on queuing and sophisticated scheduling in routers. In this paper, we address the issue of managing traffic flows with different priorities. In our reference model, incoming packets are first classified based on their priority, placed into different queues with different capacities, and then multiplexed onto one router link. The fuzzy nature of the information on Internet traffic makes this problem particularly suited to fuzzy methodologies. We propose a new solution that employs a fuzzy inference system to dynamically and efficiently schedule these priority queues. The fuzzy rules are derived to minimize the selected Lyapunov function. Simulation experiments show that the proposed fuzzy scheduling algorithm outperforms the popular Weighted Round Robin (WRR) queue scheduling mechanism.

• Journal of Internet Computing and Services
• /
• v.8 no.5
• /
• pp.125-132
• /
• 2007

Transmission mechanisms that include an available bandwidth estimation algorithm and a packet loss differentiation scheme, in general, exhibit higher TCP performance in wireless networks. TCP New Jersey, known as the best existing scheme in terms of goodput, improves wireless TCP performance using the available bandwidth estimation at the sender and the congestion warning at intermediate routers. Although TCP New Jersey achieves 17% and 85% improvements in goodput over TCP Westwood and TCP Reno, respectively, we further improve TCP New Jersey by exploring improved available bandwidth estimation, retransmission timeout, and recovery mechanisms. Hence, we propose TCP New Jersey PLUS (shortly TCP NJ+), showing that under 1% packet loss rate, it outperforms 3% by TCP New Jersey and 5% by TCP Wes1wood. In 5% packet loss rate, a characteristic of high bit-error-rate wireless network, it outperforms other TCP variants by 19% to 104% in terms of goodput even when the network is in bi-directional congestion.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.2 no.6
• /
• pp.333-351
• /
• 2008

Transmission control protocols have to overcome common problems in wireless networks. TCP employing both packet loss discrimination mechanism and available bandwidth estimation algorithm, known as the good existing solution, shows significant performance enhancement in wireless networks. For instance, TCP New Jersey which exhibits high throughput in wireless networks intends to improve TCP performance by using available bandwidth estimation and congestion warning. Even though it achieves 17% and 85% improvements in terms of goodput over TCP Westwood and TCP Reno, respectively, we further improve it by exploring maximized available bandwidth estimation, handling bit-error-rate error recovery, and effective adjustment of sending rate for retransmission timeout. Hence, we propose TCP NJ+, showing that for up to 5% packet loss rate, it outperforms other TCP variants by 19% to 104% in terms of goodput when the network is in bi-directional background traffic.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.4
• /
• pp.429-435
• /
• 2007

A number of different alternative actuation methods have been under active development for some specific applications where the traditional electromechanical actuators are difficult to apply. Recently, many of these substitutes are trying to employ new smart materials like electroactive polymers. However most of the polymeric materials are flexible and vulnerable so that they normally can not sustain external forces. Although the materials have shown a good potential to be used for alternative actuation mechanisms, no tangible industrial application is yet presented because of the reason. A conceptual design for a rectilinear motion actuator using dielectric elastomer is presented in this article. The introduced design concept might enable to produce fairly controllable rectilinear motions for various applications and the presented prototype actuator system is fully packaged in a small unit and controlled by a standard communication interface.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.8
• /
• pp.3753-3771
• /
• 2016

Increasingly, numerous and various Internet-capable devices are connected in end user networks, such as a home network. Most devices use the combination of TCP and 802.11 DCF as a system platform, but whereas some devices such as a streaming video persistently generate traffic, others such as a motion sensor do so only intermittently with lots of pauses. This study addresses the issue of performance in this heterogeneous traffic wireless LAN environment from the perspective of fairness. First, instantaneous fairness is introduced as a notion to indicate how immediately and how closely a user obtains its fair share, and a new time-based metric is defined as an index. Second, extensive simulation experiments have been made with TCP Reno, Vegas, and Westwood to determine how each TCP congestion control corresponds to the instantaneous fairness. Overall, TCP Vegas yields the best instantaneous fairness because it keeps the queue length shorter than the other TCPs. In the simulations, about 60% of a fair share of the effective user bandwidth is immediately usable in any circumstance. Finally, we introduce two simple strategies for adjusting TCP congestion controls to enhance instantaneous fairness and validate them through simulation experiments.