• Journal of KIISE
• /
• v.41 no.11
• /
• pp.974-981
• /
• 2014

These days, use of multimedia streaming service and demand of QoS (Quality of Service) improvement have been increased because of development of network. QoS of streaming service is influenced by a jitter, delay, throughput, and loss rate. For guaranteeing these factors which are influencing QoS, the role of transport layer is very important. But existing TCP which is a typical transport layer protocol increases the size of congestion window slowly and decreases the size of a congestion window drastically. These TCP characteristic have a problem which cannot guarantee the QoS of UHD multimedia streaming service. In this paper, we propose a router buffer based congestion control method for improving the QoS of UHD streaming services. Our proposed scheme applies congestion window growth rate differentially according to a degree of congestion for preventing an excess of available bandwidth and maintaining a high bandwidth occupied. Also, our proposed scheme can control the size of congestion window according to a change of delay. After comparing with other congestion control protocols in the jitter, throughput, and loss rate, we found that our proposed scheme can offer a service which is suitable for the UDH streaming service.

• Journal of the Korea Society of Computer and Information
• /
• v.5 no.3
• /
• pp.126-131
• /
• 2000

Through many users employ TCP of which the performance has been proved in Internet, but many papers Proposed to improve TCP performance according to varying network architecture. In Particular, BWDP(bandwidth-delay Product) grew larger because of the increasing delay in satellite link and the network's speed-up. To consider these increased bandwidth-delay product, it is suggested that TCP options include Window Scale option. TimeStamp option, and PAWS. Because TCP window size should be commonly high in the network with these increased bandwidth-delay product, the multiple decrease and linear increase scheme of current TCP would cause underflow and instability within network. Then TCP performance is reduced as a result. Thus, to improve TCP congestion control algorithm in the network which has large sized window, this paper proposes the congestion control scheme that controls window size by using TimeStamp option.

• Journal of KIISE:Information Networking
• /
• v.30 no.5
• /
• pp.649-655
• /
• 2003

Network-based computing such as cluster computing requires a reliable high-speed transport protocol. TCP is a representative reliable transport protocol on the Internet, which implements many mechanisms, such as flow control, congestion control, and retransmission, for reliable packet delivery. This paper, however, finds out that Myrinet does not incur any packet losses caused by network congestion. In addition, we ascertain that Myrinet supports reliable and ordered packet delivery. Consequently, most of reliable routines implemented in TCP produce unnecessarily additional overheads on Myrinet. In this paper, we show that we can attain the reliability only by flow control on Myrinet and propose a new reliable protocol based on UDP named RUM (Reliable UDP on Myrinet) that performs a flow control. As a result, RUM achieves a higher throughput by 45% than TCP and shows a similar one-way latency to UDP.

• Journal of Internet Computing and Services
• /
• v.9 no.4
• /
• pp.21-27
• /
• 2008

TCP receivers deliver ACK packets to senders for reliable end-to-end transfer. When ACK packets are not transferred properly because of network congestion, the performance of TCP degrades. This paper proposes a reverse congestion warning mechanism and a congestion handling mechanism in heterogeneous networks with heavy background traffic in the backward direction. Help from TCP receivers or hardware such as routers and gateways other than the ACK packets themselves is not necessary. TCP senders compare the arrival intervals of ACK data passed from receivers and the difference in t imestamp values echoed by receivers. According to the simulation results using the NS-2 network simulator, the proposed scheme shows a performance elevation of 20% greater than Reno, 150% greater than New Reno, and 450% greater than Westwood, respectively, under heterogeneous networks and that the error rate of the radio link is 1% when the backward network is congested.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.1B
• /
• pp.29-40
• /
• 2004

In this paper, we propose a new early congestion detection and notification technique called QR-AQM. Unlike RED and it's variation, QR-AQM measures the total traffic rate from TCP sessions, predicts future network congestion, and determine the packet marking probability based on the measured traffic rate. By incorporating the traffic rate in the decision process of the packet marking probability, QR-AQM is capable of foreseeing future network congestion as well as terminating congestion resolution procedure in much more timely fashion than RED. As a result, simulation results show that QR-AQM maintains the buffer level within a fairly narrow range around a target buffer level that may be selected arbitrarily as a control parameter. Consequently, compared to RED and its variations, QR-AQM is expected to significantly reduce the jitter and delay variance of packets traveling through the buffer while achieving nearly identical link utilization.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.42 no.12
• /
• pp.149-158
• /
• 2005

This paper proposes a congestion avoidance algorithm which provides stable throughput and transmission rate regardless of buffer size by limiting the TCP congestion window in fixed bandwidth networks. Additive Increase, Multiplicative Decrease (AIMD) is the most commonly used congestion control algorithm. But, the AIMD-based TCP congestion control method causes unnecessary packet losses and retransmissions from the congestion window increment for available bandwidth verification when used in fixed bandwidth networks. In addition, the saw tooth variation of TCP throughput is inappropriate to be adopted for the applications that require low bandwidth variation. We present an algorithm in which congestion window can be limited under appropriate circumstances to avoid congestion losses while still addressing fairness issues. The maximum congestion window is determined from delay information to avoid queueing at the bottleneck node, hence stabilizes the throughput and the transmission rate of the connection without buffer and window control process. Simulations have performed to verify compatibility, steady state throughput, steady state packet loss count, and the variance of congestion window. The proposed algorithm can be easily adopted to the sender and is easy to deploy avoiding changes in network routers and user programs. The proposed algorithm can be applied to enhance the performance of the high-speed access network which is one of the fixed bandwidth networks.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.3 s.345
• /
• pp.126-134
• /
• 2006

Traditional TCP implementations have the under-utilization problem in large bandwidth delay product networks especially during the startup phase. In this paper, we propose a delay-based congestion control(DCC) mechanism to solve the problem. DCC is subdivided into linear and exponential growth phases. When there is no queueing delay, the congestion window grows exponentially during the congestion avoidance period. Otherwise, it maintains linear increase of congestion window similar to the legacy TCP congestion avoidance algorithm. The exponential increase phase such as the slow-start period in the legacy TCP can cause serious performance degradation by packet losses in case the buffer size is insufficient for the bandwidth-delay product, even though there is sufficient bandwidth. Thus, the DCC uses the RTT(Round Trip Time) status and the estimated queue size to prevent packet losses due to excessive transmission during the exponential growth phase. The simulation results show that the DCC algorithm significantly improves the TCP startup time and the throughput performance of TCP in large bandwidth delay product networks.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07a
• /
• pp.574-576
• /
• 2005

유무선 통합망 환경에서 TCP는 무선망에서의 통신오류 밀 핸드오프를 혼잡으로 인한 패킷 손실로 간주하여 혼잡제어 메카니즘을 동작시키기 때문에 네트워크의 전반적인 효율을 저하시킨다. 이러한 문제점을 해결하기 위한 방안으로 무선망에서의 TCP 성능을 개선하기 위한 연구가 다양하게 진행되어 오고 있다. 그 중에서 Snoop 프로토콜은 종단간의 의미 구조를 유지하면서 무선망의 패킷 손실을 지역 재전송을 통하여 최소화하는 방식으로 통신오류에는 좋은 성능을 보여주지만, 대량의 무선 데이터 유입시 buffer overflow가 발생할 가능성이 높다. 따라서 본 논문에서는 버퍼관리 기법을 이용하여 buffer overflow 상황을 미연에 방지하여 불필요한 혼잡제어 메카니즘을 호출하는 것을 방지하면서 네트워크의 전반적인 효율을 향상시키는 방안을 제안하였다.

• The Journal of the Korea Contents Association
• /
• v.7 no.1
• /
• pp.131-138
• /
• 2007

In wired networks, packet losses mostly occur due to congestion. TCP reacts to the congestion by decreasing its congestion window, thus to reduce network utilization. In wireless networks, however, losses may occur due to the high bit-error rate of the transmission medium or due to fading and mobility. Nevertheless, TCP still reacts to packet losses according to its congestion control scheme, thus to reduce the network utilization unnecessarily. This reduction of network utilization causes the performance of TCP to decrease. In this paper, we predict packet loss by using RSS(Received Signal Strengths) on the wireless and suggest adding RSS flag bit in ACK packet of MH. By using RSS flag bit in ACK, the FH(Fixed Host) decides whether it adopt congestion control scheme or not for the maximum throughput. The result of the simulation by NS-2 shows that the proposed mechanism significantly increases sending amount and receiving amount by 40% at maximum.

• The KIPS Transactions:PartC
• /
• v.11C no.2
• /
• pp.235-244
• /
• 2004

TCP Vegas version provides better performance and more stable services than TCP Tahoe and Reno versions, which are widely used in the current Internet. However, in the situation where TCP Vegas and Reno share the bottleneck link, the performance of TCP Vegas is much smaller than that of TCP Reno. This unfairness is due to the difference of congestion control mechanisms of each TCP use. Several studies have been executed in order to solve this unfairness problem. In this paper, we analyze the minimum window size to maintain the maximum TCP performance of link bandwidth. In addition, we propose an algorithm which maintains the TCP performance and improves fairness by selective packet drops in order to allocate proper window size of each TCP connections. To evaluate the performance of the proposed algorithm, we have measured the number of data bytes transmitted between end-to-end systems by each TCP connections. The simulation results show that the proposed algorithm maintains the maximum TCP performance and improves the fairness.