B-ISDN is expected to support a variety of services, each with its own traffic characteristics and quality-of-service requirements. Such diversity, however, has created new congestion control problems, some of which could be alleviated by a traffic-prediction scheme. The paper investigates the applicability of artificial neural networks for traffic prediction in broadband networks. Recent work has indicated that such prediction is possible, as the neural networks are able to learn a complex mapping between past and future arrivals. Such work, however, has been based on the use of artificially generated traffic, and by definition the past and future arrivals are related. Real traffic is considered and it is shown that prediction is possible for certain traffic types but not for others. It is demonstrated that simple linear regression prediction techniques perform equally as well as do neural networks.
References
-
-
1)
-
Bellcore: ‘Star Wars’ JPEG encoded version.Available from ftp.bellcore.com/pub/vbr.video.trace/.
-
2)
-
Edwards, T., Tansley, D.S.W., Frank, R.J., Davey, N.: `Traffic trends analysis using neural networks', Proceedings of the international workshop on Applicationsof neural networks to telecommunications, 1997, p. 157–164.
-
3)
-
R. Drossu ,
T.V. Lakshman ,
Z. Obradovic ,
C. Raghavendra
.
(1995)
Single and multiple framevideo traffic prediction using neural network models, Computer networks architectures and applications.
-
4)
-
Tarraf, A.A., Habib, I.W., Saadawi, T.N.: `Neural networks for ATM multimedia traffic prediction', Proceedings of the international workshop on Applicationsof neural networks to telecommunications, 1993, 1, p. 85–91.
-
5)
-
Rose, O.: `Statistical properties of MPEG video traffic and their impact on trafficmodelling in ATM systems', Proceedings of 20th annual conference on Local computernetworks, 1995, p. 397–406.
-
6)
-
University of Wuerzburg: ananymous FTP. Available from ftp-info3.informatik.uni-wuerzburg.de/pub/MPEG/.
-
7)
-
S. Haykin
.
Neural networks: a comprehensive foundation.
-
8)
-
P.-R. Chang ,
J.-T. Hu
.
Optimal nonlinear adaptive prediction and modeling of MPEGvideo in ATM networks using pipelined recurrent neural networks.
IEEE J. Sel. Areas Commun.
,
6 ,
1087 -
1100
-
9)
-
B. Maglaris ,
D. Anastassiou ,
P. Sen ,
G. Karlsson ,
J.D. Robbins
.
Performance modelsof statistical multiplexing in packet video communications.
IEEE Trans. Commun.
,
7 ,
834 -
843
-
10)
-
Yu, E.S., Chen, C.Y.R.: `Traffic prediction using neural networks', Proceedings of IEEE Globecom '93, 1993, p. 991–995.
-
11)
-
Ott, T.J., Lakshman, T.V., Tabatabai, A.: `A scheme for smoothing delay sensitivetraffic offered to ATM networks', Proceedings of IEEE Infocom '92, 1992, p. 776–785.
-
12)
-
Garrett, M., Willinger, W.: `Analysis, modelling and generation of self-similar VBRvideo traffic', Proceedings of ACM SIGCOMM, August 1994, London, p. 269–280.
-
13)
-
Bromirski, M., Lobejko, W.: `Neural prediction of self-similar ATM traffic', Proceedings of 5th IFIP workshop on Performance modellingand evaluation of A TM networks, 1997, p. 02/1–0.2/7.
-
14)
-
K. Hornik ,
M. Stinchcombe ,
H. White
.
Multilayer feedforward networks are universal approximators.
Neural Netw.
,
359 -
366
-
15)
-
W.M. Moh ,
M.-J. Chen ,
N.-M. Chu ,
C.-D. Liao
.
Traffic prediction and dynamicbandwidth allocation over ATM: a neural network approach.
Comput. Commun.
,
8 ,
563 -
571
-
16)
-
E.A. Wan ,
A.S. Weigend ,
N.A. Gershenfeld
.
(1994)
Time series prediction by using a connectionist network with internaldelay lines, Time series prediction: forecasting the future and understanding thepast.
http://iet.metastore.ingenta.com/content/journals/10.1049/ip-com_20000146
Related content
content/journals/10.1049/ip-com_20000146
pub_keyword,iet_inspecKeyword,pub_concept
6
6