MULTIMETRIC NETWORK TOMOGRAPHY

Muhammad H. Raza, Bill Robertson, William J. Phillips, Jacek Ilow

2010

Abstract

We introduce a novel concept of multiple metric network tomography in this paper. The conventional network tomography observes a single parameter directly and infers another parameter indirectly from the the directly measured parameter. We consider observing two parameters (packet loss rate (PLR) and path delays) directly and use both of these parameters to infer a single parameter indirectly. We applied a variation of NTF1 model of non negative tensor factorization (NTF) for this purpose and estimated link delay. Simulation results should show a better correlation between the estimated and measured link delays when duplex of metrics is used as compared to using only the path level link delays for estimating the link delays on a test bed.

References

  1. Bhamidi, S., Rajagopal, R., and Roch, S. (2006). Network delay inference from additive metrics. In Arxiv preprint math/0604367.
  2. Castro, R., Coates, M., Liang, G., Nowak, R., and Yu, B. (2004). Network tomography: Recent developments. In Statistical Science. Volume 9, 499-517.
  3. Cichocki, A., Zdunek, R., Phan, A. H., , and ichi Amari, S. (2009). Nonnegative Matrix and Tensor Factorizations: Applications to Exploratory Multi-way Data Analysis and Blind Source Separation. Wiley.
  4. Coates, M. and Nowak, R. (2001). Network tomography for internal delay estimation. In 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing, Proceedings (ICASSP'01). IEEE.
  5. Hoyer, P. (2004). Non-negative matrix factorization with sparseness constraints. In The Journal of Machine Learning Research. MIT Press Cambridge, MA, USA, Volume 5, 1457-1469.
  6. Moon, S., Kurose, J., Skelly, P., and Towsley, D. (1998). Correlation of packet delay and loss in the internet. In Univ. Massachusetts, Amherst, MA, Tech. rep. CMPSCI.
  7. Vardi, Y. (1996). Network tomography: Estimating sourcedestination traffic intensities from link data. In Journal of the American Statistical Association. Volume 91, 365-377.
Download


Paper Citation


in Harvard Style

H. Raza M., Robertson B., J. Phillips W. and Ilow J. (2010). MULTIMETRIC NETWORK TOMOGRAPHY . In Proceedings of the International Conference on Data Communication Networking and Optical Communication Systems - Volume 1: DCNET, (ICETE 2010) ISBN 978-989-8425-25-6, pages 80-84. DOI: 10.5220/0002982000800084


in Bibtex Style

@conference{dcnet10,
author={Muhammad H. Raza and Bill Robertson and William J. Phillips and Jacek Ilow},
title={MULTIMETRIC NETWORK TOMOGRAPHY},
booktitle={Proceedings of the International Conference on Data Communication Networking and Optical Communication Systems - Volume 1: DCNET, (ICETE 2010)},
year={2010},
pages={80-84},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002982000800084},
isbn={978-989-8425-25-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Data Communication Networking and Optical Communication Systems - Volume 1: DCNET, (ICETE 2010)
TI - MULTIMETRIC NETWORK TOMOGRAPHY
SN - 978-989-8425-25-6
AU - H. Raza M.
AU - Robertson B.
AU - J. Phillips W.
AU - Ilow J.
PY - 2010
SP - 80
EP - 84
DO - 10.5220/0002982000800084