DocumentCode :
2518352
Title :
Weight-based fair intelligent bandwidth allocation for rate adaptive video traffic
Author :
Yu, Xiaomei ; Hoang, Doan B. ; Feng, David D.
Author_Institution :
Basser Dept. of Comput. Sci., Sydney Univ., NSW, Australia
fYear :
2000
fDate :
2000
Firstpage :
486
Abstract :
We present a weight-based network bandwidth sharing algorithm, fair intelligent bandwidth allocation (FIBA) for transporting rate adaptive video traffic using feedback, and report on its performance. Through simulations the following results are obtained. The algorithm is capable of allocating bandwidth fairly for the MCR plus weighted allocation fairness criteria and MCR plus allocation proportional to PCR fairness criteria, among competitive, rate-adaptive video sources. It is able to reallocate smoothly when there are renegotiations of the minimum guaranteed cell rate, weight, or peak cell rate by some connections, or when a new connection is admitted, or when a connection is throttled somewhere earlier along the connection path. The algorithm prevents congestion, especially during the initial periods when buffer queues can build up significantly
Keywords :
adaptive systems; bandwidth allocation; buffer storage; feedback; queueing theory; telecommunication congestion control; telecommunication traffic; visual communication; MCR; PCR fairness criteria; buffer queues; congestion avoidance; connection path; feedback; minimum guaranteed cell rate; network admission control; peak cell rate; rate adaptive video traffic; rate-adaptive video sources; renegotiations; simulations; weight-based fair intelligent bandwidth allocation; weight-based network bandwidth sharing algorithm; weighted allocation fairness criteria; Bandwidth; Channel allocation; Computational modeling; Computer science; Feedback; Intelligent networks; Switches; Telecommunication traffic; Traffic control; Video sharing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Networks, 2000. (ICON 2000). Proceedings. IEEE International Conference on
Print_ISBN :
0-7695-0777-8
Type :
conf
DOI :
10.1109/ICON.2000.875839
Filename :
875839
Link To Document :
بازگشت