Fast parameter estimation in loss tomography for networks of general topology

Deng, Ke; Li, Yang; Zhu, Weiping; Liu, Jun S.

doi:10.1214/15-aoas883

Cited by 5 publications

(4 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, more attention has been accorded to the applicability of multicast probing with general network topologies. In [10] and [11], the authors propose to use overlapping multicast trees to cover the network topology. Different Maximum Likelihood Estimators (MLE) are proposed to infer the loss rates in the intermediate links from the measurements performed over the multicast trees.…”

Section: A Metrics Inference With Multicast Probingmentioning

confidence: 99%

Unicast Inference of Additive Metrics in General Network Topologies

Rahali

Sanner

Rubino

2019

2019 IEEE 27th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS

View full text Add to dashboard Cite

Internet tomography studies the inference of the internal network performances from end-to-end measurements. Unicast probing can be advantageous for such monitoring solutions due to the wide support of unicast and the easy deployment of unicast probing paths. In this work, we propose two statistical generic methods for the inference of additive metrics using unicast probing. Our solutions give more flexibility in the choice of the collection points placement, the probed paths and they are not limited to specific topologies. Firstly, we propose the k-paths method that extends the applicability of a previously proposed solution called Flexicast for tree topologies. It is based on the Expectation-Maximization (EM) algorithm which is characterized by high computational and memory complexities. Secondly, we propose the Evolutionary Sampling Algorithm (ESA) that enhances the accuracy and the computing time but following a different approach.

show abstract

Section: A Metrics Inference With Multicast Probingmentioning

confidence: 99%

Unicast Inference of Additive Metrics in General Network Topologies

Rahali

Sanner

Rubino

2019

2019 IEEE 27th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS

View full text Add to dashboard Cite

show abstract

“…Despite this, it is not clear whether the estimate obtained the EM algorithm is the maximum likelihood estimate since there is no likelihood equation. Recently, [21] shows the EM algorithm used in [2] has its own weaknesses in terms of computation complexity. Then, an revised EM algorithm is proposed to replace the origin [21].…”

Section: A Related Workmentioning

confidence: 99%

“…Recently, [21] shows the EM algorithm used in [2] has its own weaknesses in terms of computation complexity. Then, an revised EM algorithm is proposed to replace the origin [21].…”

Section: A Related Workmentioning

confidence: 99%

“…However, some of the Weiping Zhu is with University of New South Wales, Australia results received from simulations or experiments are contradictory to each other. For example, the criticism of the MVWA in [2], [21] are not justified since there is a lack of evidence in theory to support the criticism. We argue in this paper that although there are some practical issues on the way of implementing the MVWA, the philosophy of the MVWA is sound and can be used as a reference to evaluate other estimators.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Loss Tomography in General Topology

Zhu

2016

Preprint

Self Cite

View full text Add to dashboard Cite

Although there are a few works reported in the literature considering loss tomography in the general topology, there is few well established result since all of them rely either on simulations or on experiments that have many random factors affecting the outcome. To improve the situation, we address a number of issues in this paper that include a maximum likelihood estimator (MLE) for the general topology, the statistical properties of the MLE, the statistical properties of a frequently referred estimator called the moving variance and weighted average (MVWA), and a renewed MVWA that removes the restriction of knowing variance in advance from the MVWA. The statistical properties covers minimum-variance unbiasedness, efficiency, and variances of the estimates obtained by the estimators. Given the properties, we can evaluate the estimators without the need of simulations.To verify the properties, a simulation study is conducted that confirms the accuracy of the findings.

show abstract

Statistical Methods in Network Traffic

Hazelton

2018

Wiley StatsRef: Statistics Reference Online

View full text Add to dashboard Cite

The analysis of network traffic data is of considerable interest for those working with computer networks and road transport systems. This article provides a brief overview of the modeling frameworks commonly used for describing network traffic. It covers stochastic models to describe variation in patterns of traffic flow in time, including a comparison of methods commonly used for electronic and road networks. Statistical inference for traffic models can be challenging, because the most readily available data are traffic counts observed at various network locations, but these typically provide only indirect information about the quantities of interest (e.g., traffic rates on routes through the network). We discuss methods of network tomography that address this kind of problem. Finally, we look at stochastic traffic assignment models (primarily for road traffic systems), which seek to describe the route flow dynamics of systems in which travelers attempt to minimize their own travel time.

show abstract

Fast parameter estimation in loss tomography for networks of general topology

Cited by 5 publications

References 42 publications

Unicast Inference of Additive Metrics in General Network Topologies

Unicast Inference of Additive Metrics in General Network Topologies

Loss Tomography in General Topology

Statistical Methods in Network Traffic

Contact Info

Product

Resources

About