TFRC Fairness Improving Using the Dynamic Loss Rate Measurement

In the recent years, demands of real-time audio/video stream applications have been increased. Many researches make progress actively for the new demands, and TFRC (TCP-friendly rate control) is one of the main stream in this development. This transport protocol which depends on TCP throughput equation, have been improved smoothness in transporting multimedia data and fairness with TCP. In this paper, we proposed a new loss rate calculation method that changes the weight depending on the network state as a core of the TFRC protocol. In the original TFRC method, the loss rate calculation does not consider the network state and the weight is unchanged. As a result, in a heavy-congested network, the fairness of TFRC falls off rapidly and it makes network state worse. The proposed method gives more weight to the latest loss interval, when the loss interval has decreased for 5 intervals or more in a row. In the original method, the latest loss interval is only assigned 16% of the total weight, regardless of the current congestion level. The proposed method gives more weight to the recent loss interval during the congested period. When the proposed method is used, the latest loss interval is assigned 61% of the total weight in calculating the loss rate. In a congested network, TCP's normalized throughput decreases less than 0.8 with the original method, but the proposed method can maintain the throughput more than 0.8 and enables the TCP to exist along with the TFRC.

[1]  Deborah Estrin,et al.  RAP: An end-to-end rate-based congestion control mechanism for realtime streams in the Internet , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).

[2]  Mark Handley,et al.  A Comparison of Equation-Based and AIMD Congestion Control , 2000 .

[3]  R. Rejaie RAP: End-to-end Rate Based Control for Real Time Streams in the Internet , 2003, INFOCOM 1999.

[4]  Matthew Mathis,et al.  The stationary behavior of ideal TCP congestion avoidance , 1996 .

[5]  Martin Mauve,et al.  A survey on TCP-friendly congestion control , 2001, IEEE Netw..

[6]  Donald F. Towsley,et al.  Modeling TCP throughput: a simple model and its empirical validation , 1998, SIGCOMM '98.

[7]  Min Sik Kim,et al.  Transient behaviors of TCP-friendly congestion control protocols , 2003, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[8]  Matthew Mathis,et al.  The macroscopic behavior of the TCP congestion avoidance algorithm , 1997, CCRV.

[9]  Jon Crowcroft,et al.  Congestion control mechanisms and the best effort service model , 2001, IEEE Netw..

[10]  Jörg Widmer,et al.  TCP Friendly Rate Control (TFRC): Protocol Specification , 2003, RFC.

[11]  QUTdN QeO,et al.  Random early detection gateways for congestion avoidance , 1993, TNET.

[12]  Mark Handley,et al.  Equation-based congestion control for unicast applications , 2000, SIGCOMM.

[13]  Jongwon Lee,et al.  ATFRC: Adaptive TCP Friendly Rate Control Protocol , 2003, ICOIN.

[14]  Sally Floyd,et al.  Promoting the use of end-to-end congestion control in the Internet , 1999, TNET.