Application of Support Vector Machine to Mobile Communications in Telephone Traffic Load of Monthly Busy Hour Prediction

Author

Han, Rui ; Jia, Zhenhong ; Qin, Xizhong ; Chang, Chun ; Wang, Hao

Author_Institution

Coll. of Inf. Sci. & Eng., Xinjiang Univ., Urumqi, China

Volume

3

fYear

2009

fDate

14-16 Aug. 2009

Firstpage

349

Lastpage

353

Abstract

Telephone traffic of busy hour is one of indicators of load capacity of telecommunication network, which has a significant meaning to dilate and modify the network. A good performance of predicting the monthly busy hour traffic load is cared about by the mobile operators. As a promising learning theory, support vector machine (SVM) has been studied and applied in a wide area, such as financial markets and weather forecast. In this paper, we use SVM to forecast monthly busy hour traffic load of two regions in Xinjiang. A good result has been achieved via an improved grid search method for the search of hyper-parameter of SVM.

Keywords

grid computing; mobile communication; mobile computing; support vector machines; telephone traffic; time series; SVM; busy hour traffic load prediction; grid search method; learning theory; mobile communication; support vector machine; telecommunication network load capacity; telephone traffic load; Artificial neural networks; Economic forecasting; Electronic mail; Load forecasting; Mobile communication; Search methods; Support vector machines; Telecommunication traffic; Telephony; Weather forecasting; SVM; busy hour load forecasting; improved grid search;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation, 2009. ICNC '09. Fifth International Conference on

Conference_Location

Tianjin

Print_ISBN

978-0-7695-3736-8

Type

conf

DOI

10.1109/ICNC.2009.96

Filename

5366848