Results 1 
6 of
6
The role of PASTA in network measurement
 Comput. Commun. Rev
, 2006
"... property applicable to many stochastic systems. In active probing, PASTA is invoked to justify the sending of probe packets (or trains) at Poisson times in a variety of contexts. However, due to the diversity of aims and analysis techniques used in active probing, the benefits of Poissonbased measu ..."
Abstract

Cited by 16 (6 self)
 Add to MetaCart
property applicable to many stochastic systems. In active probing, PASTA is invoked to justify the sending of probe packets (or trains) at Poisson times in a variety of contexts. However, due to the diversity of aims and analysis techniques used in active probing, the benefits of Poissonbased measurement, and the utility and role of PASTA, are unclear. Using a combination of rigorous results and carefully constructed examples and counterexamples, we map out the issues involved and argue that PASTA is of very limited use in active probing. In particular, Poisson probes are not unique in their ability to sample without bias. Furthermore, PASTA ignores the issue of estimation variance and the central need for an inversion phase to estimate the quantity of interest based on what is directly observable. We give concrete examples of when Poisson probes should not be used, explain why, and offer initial guidelines on suitable alternative sending processes. Index Terms—Active probing, network measurement, Nonintrusive
Measurement of packet loss probability by optimal design of packet probing experiments
, 2008
"... Packet level measurement is now critical to many aspects of broadband networking, e.g. for guaranteeing Service Level Agreements, facilitating measurementbased admission control algorithms, and performing network tomography. Because it is often impossible to measure all the data passing across a ne ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
Packet level measurement is now critical to many aspects of broadband networking, e.g. for guaranteeing Service Level Agreements, facilitating measurementbased admission control algorithms, and performing network tomography. Because it is often impossible to measure all the data passing across a network, the most widely used method of measurement works by injecting probe packets. The probes provide samples of the packet loss and delay, and from these samples the loss and delay performance of the traffic as a whole can be deduced. However measuring performance like this is prone to errors. Recent work has shown that some of these errors are minimized by using a gamma renewal process as the optimal pattern for the time instants at which to inject probes. This leaves the best rate at which to inject probes as the key unsolved problem, and we address this here by using the statistical principles of the Design of Experiments. The experimental design approach allows us 1 to treat packet level measurements as numerical experiments that can be designed optimally. Modelling the overflow of buffers as a 2state Markov chain, we deduce the system’s likelihood function, and from this we develop a technique (using the Fisher information matrix) to determine the upperbound on the optimal rate of probing. A generalization of this method accounts for the effect of the probed observations interfering with the experiment. Our numerical results focus on VoIP traffic, allowing us to show how this methodology would be used in practice. One application of this is in measurementbased admission control algorithms, where our technique can be used to provide an upperbound on the rate at which probes should be injected to monitor the loss performance of the target network, prior to making an admit / don’t admit decision. 1
Towards informative statistical flow inversion
 CoRR
"... A problem which has recently attracted research attention is that of estimating the distribution of flow sizes in internet traffic. On high traffic links it is sometimes impossible to record every packet. Researchers have approached the problem of estimating flow lengths from sampled packet data in ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
A problem which has recently attracted research attention is that of estimating the distribution of flow sizes in internet traffic. On high traffic links it is sometimes impossible to record every packet. Researchers have approached the problem of estimating flow lengths from sampled packet data in two separate ways. Firstly, different sampling methodologies can be tried to more accurately measure the desired system parameters. One such method is the sampleandhold method where, if a packet is sampled, all subsequent packets in that flow are sampled. Secondly, statistical methods can be used to “invert ” the sampled data and produce an estimate of flow lengths from a sample. In this paper we propose, implement and test two variants on the sampleandhold method. In addition we show how the sampleandhold method can be inverted to get an estimation of the genuine distribution of flow sizes. Experiments are carried out on real network traces to compare standard packet sampling with three variants of sampleandhold. The methods are compared for their ability to reconstruct the genuine distribution of flow sizes in the traffic.
Techniques for flow inversion on sampled data
"... Abstract—The distribution of flow sizes is a quantity of interest fundamental to traffic engineering and network modelling and only likely to become more important in the future. The recovery of the flowlength distribution from (sampled) packet data is referred to as flowinversion. Traditional pac ..."
Abstract
 Add to MetaCart
Abstract—The distribution of flow sizes is a quantity of interest fundamental to traffic engineering and network modelling and only likely to become more important in the future. The recovery of the flowlength distribution from (sampled) packet data is referred to as flowinversion. Traditional packet sampling methods cause distortions in a recovered distribution of flowlength. We propose an improved method for inverting data sampled using the technique known as sampleandhold. We show that the technique improves upon existing inversion techniques illustrated using both real and artificial data sets. The technique described may have applications to other inversion problems. I.
1 Consistent estimation of nonbandlimited spectral density from uniformly spaced samples
, 906
"... Abstract—In the matter of selection of sample time points for the estimation of the power spectral density of a continuous time stationary stochastic process, irregular sampling schemes such as Poisson sampling are often preferred over regular (uniform) sampling. A major reason for this preference i ..."
Abstract
 Add to MetaCart
Abstract—In the matter of selection of sample time points for the estimation of the power spectral density of a continuous time stationary stochastic process, irregular sampling schemes such as Poisson sampling are often preferred over regular (uniform) sampling. A major reason for this preference is the wellknown problem of inconsistency of estimators based on regular sampling, when the underlying power spectral density is not bandlimited. It is argued in this paper that, in consideration of a large sample property like consistency, it is natural to allow the sampling rate to go to infinity as the sample size goes to infinity. Through appropriate asymptotic calculations under this scenario, it is shown that the smoothed periodogram based on regularly spaced data is a consistent estimator of the spectral density, even when the latter is not bandlimited. It transpires that, under similar assumptions, the estimators based on uniformly sampled and Poissonsampled data have about the same rate of convergence. Apart from providing this reassuring message, the paper also gives a guideline for practitioners regarding appropriate choice of the sampling rate. Theoretical calculations for large samples and MonteCarlo simulations for small samples indicate that the smoothed periodogram based on uniformly sampled data have less variance and more bias than its counterpart based on Poisson sampled data. Index Terms—spectral estimation, periodogram, regular sampling, Poisson sampling, consistency, rates of convergence. I.
Problèmes inverses dans les réseaux Inverse problems in Networks
, 2013
"... (Ecole doctorale: Informatique, télécommunications et électronique (ED 130)) présentée par ..."
Abstract
 Add to MetaCart
(Ecole doctorale: Informatique, télécommunications et électronique (ED 130)) présentée par