Title :
Study on network traffic prediction techniques
Author :
Feng, Huifang ; Shu, Yantai
Author_Institution :
Dept. of Comput. Sci., Tianjin Univ., China
Abstract :
We briefly describe a number of traffic predictors (such as ARIMA, FARIMA, ANN and wavelet-based predictors) and analyze their computational complexity. We compare their performance with MSE, NMSE and computational complexity by simulating the predictors on four wireless network traffic traces and decide the most suitable network traffic predictor based on acceptable performance and accuracy.
Keywords :
autoregressive moving average processes; computational complexity; neural nets; radio networks; telecommunication computing; telecommunication traffic; wavelet transforms; artificial neural network; computational complexity; fractional autoregressive integrated moving average; network traffic prediction techniques; wavelet-based predictors; wireless network traffic; Artificial neural networks; Channel allocation; Communication system traffic control; Computational complexity; Computer science; Measurement; Neural networks; Predictive models; Telecommunication traffic; Traffic control;
Conference_Titel :
Wireless Communications, Networking and Mobile Computing, 2005. Proceedings. 2005 International Conference on
Print_ISBN :
0-7803-9335-X
DOI :
10.1109/WCNM.2005.1544219
