DocumentCode :
1733733
Title :
PDF based congestion control in ATM networks
Author :
Unteregelsbacher, Erich ; Mouft, H.T.
Author_Institution :
Dept. of Electr. Eng., Queen´´s Univ., Kingston, Ont., Canada
fYear :
1991
Firstpage :
211
Abstract :
The authors describe the problem of congestion control by looking at the probability density function (PDF) of the traffic intensity. It is pointed out that the various congestion control methods found in the literature can be understood as attempts to modify the shape of the PDF so that its average value increases. This has the side effect of introducing lower service traffic classes. It is proposed that the network only carry one traffic class providing the best service the network can offer. Bandwidth utilization can be increased by allowing the user to send `marked´ cells which the network will only carry if no normal cells are available. The bandwidth to be allocated statistically can be calculated from the PDF, which in turn can be calculated from the PDFs of the individual calls. No (minimal) buffering is required
Keywords :
probability; telecommunication networks; telecommunication traffic; telecommunications control; time division multiplexing; ATM networks; PDF; marked cells; probability density function; service traffic classes; statistical bandwidth allocation; traffic intensity; Bandwidth; Communication system traffic control; Convolution; Gaussian distribution; Intelligent networks; Probability density function; Proposals; Shape control; Switches; Tail;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Communications, 1991. ICC '91, Conference Record. IEEE International Conference on
Conference_Location :
Denver, CO
Print_ISBN :
0-7803-0006-8
Type :
conf
DOI :
10.1109/ICC.1991.162361
Filename :
162361
Link To Document :
بازگشت