Robust Identification of Shared Losses Using End-to-End Unicast Probes
Robust Identification of Shared Losses Using End-to-End Unicast Probes
Maximum pseudo likelihood estimation in network tomography
IEEE Transactions on Signal Processing
IEEE Transactions on Signal Processing
Unicast-based inference of network link delay distributions with finite mixture models
IEEE Transactions on Signal Processing
Multicast-based inference of network-internal loss characteristics
IEEE Transactions on Information Theory
Using Bayesian network on network tomography
Computer Communications
Network tomography: identifiability and Fourier domain estimation
IEEE Transactions on Signal Processing
Bottom up algorithm to identify link-level transition probability
ICCNMC'05 Proceedings of the Third international conference on Networking and Mobile Computing
Hi-index | 0.25 |
Loss tomography, as a key component of network tomography, aims to obtain the loss rate of each link in a network by end-to-end measurements. If knowing the loss model of a link, we, in fact, deal with a parametric estimate problem with incomplete data. Maximum likelihood estimates are often used in this situation to identify the unknown parameters in the loss model. Almost all methods proposed so far rely on the iterative approximation to identify the parameters that requires a long execution time. In addition, the parameters identified by those methods may not be the true values of those parameters since the iterative procedure may trap into a local maximum. In this paper, we propose an estimate that is based on the correlation between a link and its sibling brothers to identify the loss rate of the link. The proposed method, instead of using an iterative approach to approximate the maximum, employs a bottom-up approach to identify the loss rates of the links of a network. Comparing to the previous methods, the proposed method is simple and fast because it is an analytical solution.