Multi-scale traffic matrix estimation algorithm

The traffic matrix is one of the crucial inputs in many traffic engineering and network planning tasks, but it is usually impossible to directly measure traffic matrices. So, it is an important research topic to infer traffic matrix by reasonably modeling, and incorporating the limited empirical information. Of the proposed methods, Kalman Filtering method is a more efficient and accurate method than many others. But Kalman Filter method is the linear estimation area, this limit of the linear nature make it impossible to estimate the self-similar, long range dependence and multi-scale characteristic accurately. Multi-fractal wavelet model can handle the nonlinear nature of the network traffic efficiently. Considering the linear and nonlinear nature, this paper proposed the multi-scale traffic matrix estimation algorithm based on the incorporation of the Kalman Filter and wavelet analysis.