Multiscale nature of network traffic

Patrice Abry, Richard Baraniuk, Patrick Flandrin, Rudolf Riedi, Darryl Veitch

Research output: Contribution to journalArticlepeer-review

247 Scopus citations


The complexity and richness of teletraffic is shown to be well matched by the multiscale analysis and modeling frameworks of self-similarity, long-range dependence, fractals, multifractals, and infinitely divisible cascades. These frameworks not only allow to confirm and formalize the presence of multiscale behavior in traffic, but also point to possible causes of multiscale structure in the physical networking infrastructure. Whatever choice of the framework, the multiscale wavelet transform provides a parsimonious and efficient domain for processing.

Original languageEnglish (US)
Pages (from-to)28-46
Number of pages19
JournalIEEE Signal Processing Magazine
Issue number3
StatePublished - May 2002

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Applied Mathematics


Dive into the research topics of 'Multiscale nature of network traffic'. Together they form a unique fingerprint.

Cite this