• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.1
• /
• pp.139-145
• /
• 2016

Dumbbell is the most basic topology that can be used in almost all kind of network experiment within it or just by little expansion. While Transmission Control Protocol TCP is the basic protocol that is used for the connectivity among networks and stations. TCP major and basic goal is to provide path and services to different applications for communication. For that reason TCP has to transfer a lot of data through a communication medium that cause serious congestion problem. To calculate the congestion problem, different kind of pre-cure solutions are developer which are Loss Based Variant and Delay Based Variant. While LBV keep track of the data that is going to be passed through TCP protocol, if the data packets start dropping that means congestion occurrence which notify as a symptom, TCP CUBIC use LBV for notifying the loss. Similarly the DBV work with the acknowledgment procedure that is used in when data ACK get late with respect to its set data rate time, TCP COMPOUND/VAGAS are examples of DBV. Many algorithms have been purposed to control the congestion in different TCP variants but the loss of data packets did not completely controlled. In this paper, the congestion control algorithms are implemented and corresponding results are analyzed in Dumbbell topology, it is typically used to analyze the TCP traffic flows. Fairness of throughput is evaluated for different TCP variants using network simulator (NS-2).

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.1151-1154
• /
• 2006

다양한 무선 네트워킹의 출현은 유무선 혼합형태의 네트워크 망으로 변화되면서 기존의 유선 중심의 네트워크 기반으로 구현된 TCP에 많은 문제점을 가져왔다. 높은 비트오류율, 낮은 대역폭 그리고 높은 지연 등은 TCP/IP 통신을 효율적으로 지원하지 못하는 원인이 된다. 또한 BS(Base Station:기지국)에서 보낸 패킷이 MH(Mobile Node : 이동성을 갖는 노드)가 수신범위 밖으로 벗어나면서 손실로 이어질 경우 성능 저하의 원인이 되기도 한다. 본 논문에서는 MN의 ACK에 수신신호 관련 flag bit를 추가하면서 BS에서 추가된 ACK의 flag bit를 이용해서 패킷손실을 최소화하는 기법을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.05a
• /
• pp.1267-1270
• /
• 2005

센서 네트워크에서 센서 노드의 전력 소모는 최대한 줄이면서, 원격 측정 결과에 대한 가독성을 높이기 위해 계측된 데이터의 신뢰도를 높이는 게 중요하다. 본 논문에서는 계측된 데이터와 임계값간 차가 커질수록 센서 노드의 데이터 전송 주기를 짧게 하는 센서 데이터 종속적인 비주기 전송 방식을 제안하였다. 기존에 제안된 비주기 전송방식과 비교해 본 결과 전체 전송 횟수는 비슷하지만, 데이터 변화량이 큰 구간에서는 최소 2 배 이상 전송 횟수가 증가함을 확인하였다. 한편, 패킷 오류율을 고려하지 않은 상태에서 대역폭, 센서 노드 수, 패킷 크기 및 전송 주기 등 4 가지 평가 조건에 따른 TCP 와 UDP 의 성능 비교 분석 결과, ACK 패킷을 제외했을 경우 두 프로토콜 간 전송 성능 차는 거의 없는 것으로 나타났다. 즉 낮은 PER 을 갖는 센서 네트워크인 경우에는 UDP 가 적합하며, PER 이 높을 경우에는 ACK 패킷에 대한 오버헤드를 부담하더라도 계측된 데이터의 신뢰도 보장을 위해 TCP 가 적합한 것으로 나타났다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10c
• /
• pp.181-183
• /
• 2004

본 논문은 TCP에서 사용되는 혼잡 제어 방식의 문제점인 전송 시간에 대한 전송률 편중 현상을 완화시키기 위한 비례 증가 혼잡 제어 방식을 제안한다. TCP는 네트워크의 혼잡 상황을 제어하기 위하여 ACK 기반의 혼잡 윈도우 조정 방식을 사용한다. 그러나 기존 TCP에서 사용되는 혼잡 제어 방식은 그 자체적으로 네트워크의 혼잡 상황을 유도하며, 전송 시간이 다른 두 개의 TCP 연결이 하나의 병목 지점을 공유할 경우, 전송 시간의 차이에 따라 전송률의 편중 현상이 나타난다. 본 논문에서는 이러한 네트워크 혼잡 상황 유발과 전송 시간의 차이에 따른 전송률 편중 현상을 완화시키기 위한 송신자 기반의 네트워크 대역폭 측정 방법 및 비례 증가 혼잡 제어 알고리즘을 제안하며, 구현 및 실험을 통해 기존 TCP의 문제점을 해결할 수 있음을 보인다.

• The KIPS Transactions:PartC
• /
• v.9C no.4
• /
• pp.513-522
• /
• 2002

Current end-to-end congestion control depends only on the information of end points (using three duplicate ACK packets) and generally responds slowly to the network congestion. This mechanism can't avoid TCP global synchronization which TCP congestion window size is fluctuated during congestion occurred and if RTT (Round Trip Time) is increased, three duplicate ACK packets is not a correct congestion signal because congestion maybe already disappeared and the host may send more packets until receive the three duplicate ACK packets. Recently there is increasing interest in solving end-to-end congestion control using active network frameworks to improve the performance of TCP protocols. ACC (Active congestion control) is a variation of TCP-based congestion control with queue management In addition traffic modifications nay begin at the congested router (active router) so that ACC will respond more quickly to congestion than TCP variants. The advantage of this method is that the host uses the information provided by the active routers as well as the end points in order to relieve congestion and improve throughput. In this paper, we model enhanced ACC, provide its algorithm which control the congestion by using information in core networks and communications between active routers, and finally demonstrate enhanced performance by simulation.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.7
• /
• pp.413-420
• /
• 2006

The speed of present-day network technology exceeds a gigabit and is developing rapidly. When using TCP/IP in these high-speed networks, a high load is incurred in processing TCP/IP protocol in a host CPU. To solve this problem, research has been carried out into TCP/IP Offload Engine (TOE). The TOE processes TCP/IP on a network adapter instead of using a host CPU; this reduces the processing burden on the host CPU. In this paper, we developed two software-based TOEs. One is the TOE implementation using an embedded Linux. The other is the TOE implementation using Lightweight TCP/IP (lwIP). The TOE using an embedded Linux did not have the bandwidth more than 62Mbps. To overcome the poor performance of the TOE using an embedded Linux, we ported the lwIP to the embedded system and enhanced the lwIP for the high performance. We eliminated the memory copy overhead of the lwIP. We added a delayed ACK and a TCP Segmentation Offload (TSO) features to the lwIP and modified the default parameters of the lwIP for large data transfer. With the aid of these modifications, the TOE using the modified lwIP shows a bandwidth of 194 Mbps.

• The KIPS Transactions:PartC
• /
• v.13C no.5 s.108
• /
• pp.635-640
• /
• 2006

The Snoop in the BS (Base Station) performs a local retransmission over wired-wireless networks to recover packet loss quickly. However, when the MH (Mobile Host) leaves the reception range of the signal, the local retransmission causes performance to degrade. In this paper, we minimize the packet loss and local retransmission caused by the unreachability from BS to MH to improve network performance. To do this, we suggest to add RSS(Received Signal Strengths) flag bit in ACK packet of MH. By using this flag bit, the BS decides whether it retransmits or not to minimize packet loss. The result of the simulation by ns-2 shows a big improvement of performance in the networks.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.5
• /
• pp.923-930
• /
• 2001

In this paper, in order to improve the performance of TCP short traffic services in wireless Internet environments, the Split-ACKs(SPACK) scheme is proposed. In wireless networks, unlike wired networks, packet losses will occur more often due to high bit error rates. Therefore, each packet loss over wireless links results in congestion control procedure of TCP being invoked at the source. This causes severe end-to-end performance degradation of TCP. In this paper, to alleviate the TCP performance, the SPACK method, split acknowledgement packets in the base station, is proposed. Using computer simulation, the performance of TCP using SPACK is analyzed and shows better performance than traditional TCP protocol.

• Journal of Communications and Networks
• /
• v.8 no.3
• /
• pp.351-359
• /
• 2006

In this article, we present a new slow-start variant, which improves the throughput of transmission control protocol (TCP) Vegas. We call this new mechanism Gallop-Vegas because it quickly ramps up to the available bandwidth and reduces the burstiness during the slow-start phase. TCP is known to send bursts of packets during its slow-start phase due to the fast window increase and the ACK-clock based transmission. This phenomenon causes TCP Vegas to change from slow-start phase to congestion-avoidance phase too early in the large bandwidth-delay product (BDP) links. Therefore, in Gallop-Vegas, we increase the congestion window size with a rate between exponential growth and linear growth during slow-start phase. Our analysis, simulation results, and measurements on the Internet show that Gallop-Vegas significantly improves the performance of a connection, especially during the slow-start phase. Furthermore, it is implementation feasible because only sending part needs to be modified.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.3
• /
• pp.523-541
• /
• 2011

In current IEEE 802.11 wireless LAN, every unicast transmission requires an ACK from the receiver for reliability, though it consumes energy and bandwidth. There have been studies to remove or reduce ACK overhead, especially for energy efficiency. However none of them are practically used now. This paper introduces a noble method of selective unacknowledged transmission, where skipping an ACK is dynamically decided frame by frame. Utilizing the fact that a multicast frame is transmitted without accompanying an ACK in 802.11, the basic unacknowledged transmission is achieved simply by transforming the destination address of a frame to a multicast address. Since removing ACK is inherently more efficient but less strict, its practical profit is dependent on traffic characteristics of a frame as well as network error conditions. To figure out the selective conditions, energy and performance implications of unacknowledged transmission have been explored. Extensive experiments show that energy consumption is almost always reduced, but performance may be dropped especially when TCP exchanges long data with a long distance node through a poor wireless link. An experiment with a well-known traffic model shows that selective unacknowledged transmission gives energy saving with comparable performance.