Design of experiments for categorical repeated measurements in packet communication networks
Design of experiments for categorical repeated measurements in packet communication networks
We study the optimal measurement of packet loss and delay in packet networks by treating all measurements as numerical experiments to which we apply the theory of the design of experiments. Specifically we seek to find the optimal times at which to inject survey (probe) packets. Our approach is to model the target node in the packet communication network, an access buffer, as a discrete-time Markov chain. Given that we may only make a limited number of observations, we present a method for optimally designing the observation times for the chain, and derive both exact and continuous optimal designs. Our results show that, for common optimality criteria, measuring at a uniform rate may not be optimal. This has significance for influencing commercial practice as uniform probing is standard. We show how our method may be generalized to Markov systems with larger state space, and describe computational methods to find optimal designs on any system which evolves according to the Markov principle.
automatic differentiation, bayesian design, d-optimal, maximum likelihood estimation
339-352
Parker, Ben M.
26c5a5ab-17b3-4d6c-ae11-abf3a2554529
Gilmour, Steven G.
984dbefa-893b-444d-9aa2-5953cd1c8b03
Schormans, John A.
1dca07b2-e968-4e58-aede-ec203f0965a8
November 2011
Parker, Ben M.
26c5a5ab-17b3-4d6c-ae11-abf3a2554529
Gilmour, Steven G.
984dbefa-893b-444d-9aa2-5953cd1c8b03
Schormans, John A.
1dca07b2-e968-4e58-aede-ec203f0965a8
Parker, Ben M., Gilmour, Steven G. and Schormans, John A.
(2011)
Design of experiments for categorical repeated measurements in packet communication networks.
Technometrics, 53 (4), .
(doi:10.1198/TECH.2011.10052).
Abstract
We study the optimal measurement of packet loss and delay in packet networks by treating all measurements as numerical experiments to which we apply the theory of the design of experiments. Specifically we seek to find the optimal times at which to inject survey (probe) packets. Our approach is to model the target node in the packet communication network, an access buffer, as a discrete-time Markov chain. Given that we may only make a limited number of observations, we present a method for optimally designing the observation times for the chain, and derive both exact and continuous optimal designs. Our results show that, for common optimality criteria, measuring at a uniform rate may not be optimal. This has significance for influencing commercial practice as uniform probing is standard. We show how our method may be generalized to Markov systems with larger state space, and describe computational methods to find optimal designs on any system which evolves according to the Markov principle.
Text
Technometrics_2011.pdf
- Version of Record
Restricted to Repository staff only
Request a copy
More information
Published date: November 2011
Keywords:
automatic differentiation, bayesian design, d-optimal, maximum likelihood estimation
Organisations:
Statistics
Identifiers
Local EPrints ID: 337609
URI: http://eprints.soton.ac.uk/id/eprint/337609
ISSN: 0040-1706
PURE UUID: 4dfbe67e-5096-4695-8d05-dfbea2c1dd3f
Catalogue record
Date deposited: 01 May 2012 10:12
Last modified: 14 Mar 2024 10:56
Export record
Altmetrics
Contributors
Author:
Ben M. Parker
Author:
Steven G. Gilmour
Author:
John A. Schormans
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics