On the Relevance of Long-Range Dependence in Network Traffic by M. Glossglauser and J. Bolot
Motivations
A modulated fluid traffic model
Special case: Truncated Pareto distribution
Covariance
Compute loss rate
Numerical results (1)
Loss rate as a function of buffer size and cutoff time
Numerical results (2)
Loss rate as a function of Hurst and marginal distribution
Numerical results (3)
Loss rate as a function of marginal distribution and buffer size
Conclusion from numerical experiments
Conclusion of the paper
Email: jins@cs.bu.edu
Home Page: http://www.cs.bu.edu/students/grads/jins/Home.html