Title :
Performing traffic pattern prediction in WCDMA networks
Author :
Zhang, Kejing ; Cuthbert, Laurie
Author_Institution :
Dept. of Electron. Eng., Queen Mary Univ. of London, London
Abstract :
Early detection and rapid resolution network congestion can considerably improve network capacity. Consequently, much research has been carried out on predicting traffic patterns in 3G networks. This paper introduces an access point centric approach that is implemented by two kinds of predictors, the WPML predictor and the order-k Markov predictor. The feasibility of this approach is evaluated, and the prediction performance of these two types of predictors is compared. Finally, we conclude that the access point centric approach is competent for both the one-step-ahead and multi-step-ahead traffic pattern prediction.
Keywords :
3G mobile communication; Markov processes; broadband networks; code division multiple access; telecommunication traffic; 3G networks; Markov predictors; WCDMA networks; multistep-ahead traffic pattern prediction; network capacity; traffic patterns; Antenna radiation patterns; Frequency; Maximum likelihood estimation; Multiaccess communication; Pattern analysis; Pattern recognition; Shape control; Telecommunication traffic; Vehicles; Wavelet analysis; Congestion control; Markov chain; Maximum likelihood estimation; Traffic pattern prediction; Weighted probability;
Conference_Titel :
Wavelet Analysis and Pattern Recognition, 2008. ICWAPR '08. International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-2238-8
Electronic_ISBN :
978-1-4244-2239-5
DOI :
10.1109/ICWAPR.2008.4635892
