One Method from LRD to SRD
Research of self-similarity has been the focus of flow prediction due to the complexity of self-similar traffic, but the LRD (long range dependence) traffic has the disadvantages of the high modeling algorithm complexity and poor precision. Therefore we tried to propose a way that can make the LRD t...
Gespeichert in:
Hauptverfasser: | , , , |
---|---|
Format: | Tagungsbericht |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Research of self-similarity has been the focus of flow prediction due to the complexity of self-similar traffic, but the LRD (long range dependence) traffic has the disadvantages of the high modeling algorithm complexity and poor precision. Therefore we tried to propose a way that can make the LRD traffic change to SRD (short range dependence) which is more simple than LRD in the field of modeling and predicting. The researchers adopted EMD (empirical mode decomposition) to decompose LRD data which would be decomposed into several IMF (intrinsic mode function) components. Then we found that IMF components had no longer self-similar property through theoretical analysis and simulation, thus people could use some SRD model to forecast traffic. |
---|---|
ISSN: | 2161-9646 |
DOI: | 10.1109/WICOM.2009.5301626 |