Title :
Probing for Loss: The Case Against Probe Trains
Author :
Baccelli, François ; Machiraju, Sridhar ; Veitch, Darryl ; Bolot, Jean
Author_Institution :
INRIA, ENS, Paris, France
fDate :
5/1/2011 12:00:00 AM
Abstract :
We assess the bias and variance of active probing measurement techniques for loss in networks. We show that loss probability is best measured using isolated probes rather than with probe `trains´, which contrasts with prior work which found that probe trains better estimate statistics of lossy periods.
Keywords :
computer networks; losses; probability; active probing measurement techniques; computer networks; isolated probes; loss probability; network loss; probe trains; statistic estimation; Correlation; Delay; Estimation; Internet; Loss measurement; Probability; Probes; NIMASTA; Probing; loss; optimal variance; probe trains;
Journal_Title :
Communications Letters, IEEE
DOI :
10.1109/LCOMM.2011.030911.101495
