Study on Traffic Prediction Algorithm Based on Multi-Scale Bi-Orthogonal Wavelet

Abstract:

The traffic prediction carried out in the communication enterprises is of great significance for the optimization of the network configuration and the improvement of the communication quality. To solve the inaccurate prediction problem under the actual situation, a traffic prediction method based on the bi-orthogonal multi-scale wavelet algorithm is developed. The process of the wavelet decomposition and reconstruction are studied, and the reconstruction results for the different scales wavelet are obtained. Take a set of the special actual samples as the object, the traffic prediction for the future dates is completed, and compared with the actual results. The results show that the relative error between the proposed traffic prediction model and the actual results is less than 10%. The bi-orthogonal multi-scale wavelet algorithm has some advantages as compared with other similar ones, which will provide the important technology means for the traffic prediction forecasting and assessing in the various types of communication enterprises.

• Periodical: Applied Mechanics and Materials (Volume 340)
• Pages: 722-726
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.340.722
• Online since: July 2013
• Authors: Yan Li, Yao Chen
• Keywords: Bi-Orthogonal Wavelets, Multi-Scale Wavelet, Network Optimization, Traffic Prediction

References

[1] Paxson V, Floyd S (1995) Wide-area traffic: The failure of poisson modeling. IEEE/ACM Trans. on Networking 3(3): 226-244.

DOI: 10.1109/90.392383

[2] Sheng M, Chuanyi J (2001) Modeling heterogeneous network traffic in wavelet domain. IEEE/ACM Trans on Networking 9(5): 634-649.

DOI: 10.1109/90.958331

[3] Wagenmakers EJ, Grunwald P, Steyvers M (2006) Accumulative prediction error and the selection of time series models. Journal of Mathematical Psychology 50(2): 149-166.

DOI: 10.1016/j.jmp.2006.01.004

[4] Willinger W, Paxson V, Riedi R H, et al (2001) Long-range dependence and data network traffic. Cambridge, MA: Birkhauser 1: 68-73.

[5] Stephane G M (1989) a theory for multiresolution signal decomposition: The wavelet representation. IEEE Tran on Pattern Analysis and Machine Intelligence 11(7): 674-693.

DOI: 10.1109/34.192463

[6] Cheng JL (2006) Wavelet neural networks with a hybrid learning approach. Journal of Information Science and Engineering 22(6): 1367-1387.

[7] Daubechies I (1988) Orthonormal bases of compactly supported wavelets. Communication on Pure and Applied Mathematics 41(11): 909-996.

DOI: 10.1002/cpa.3160410705

[8] Ortigueira MD, Serralheiro AJ (2006) A new least-squares approach to differ integration modeling. Signal Processing 86(10): 2582-2591.

DOI: 10.1016/j.sigpro.2006.02.013