China
Africa
Explore chapters and articles related to this topic
Utility-Optimized Aggregate Flow Level Bandwidth Allocation
Published in Liansheng Tan, Resource Allocation and Performance Optimization in Communication Networks and the Internet, 2017
Under a certain routing scheme, the above routing association tells us that the individual flow rate vector can determine the aggregate flow rate vector. In the case of the routing matrix A which is invertible, the individual flow rate vector and the aggregate flow rate vector are of one-to-one correspondence. That is, for a certain routing scheme, the aggregate flow rate vector can also determine the individual flow rate vector uniquely. In relation to this issue, the inference of OD byte counts from link byte counts measured at router interfaces under a fixed routing schemes has been discussed in the so-called network tomography [130], where the authors use the statistics method.
Inferring sources of substandard and falsified products in pharmaceutical supply chains
Published in IISE Transactions, 2023
Eugene Wickett, Matthew Plumlee, Karen Smilowitz, Souly Phanouvong, Victor Pribluda
Network-tomography methods infer unknown network parameters through measurements taken at a subset of network locations (Castro et al., 2004). Network tomography emerged with the internet’s rise, as transfer delay could only be measured at origins and destinations while delay at interior network links remained unknown. A frequently studied model is where is a vector of link-level measurements of a phenomenon such as traffic flow or delay, is a vector of parameters characterizing phenomena for paths between pairs of nodes, and is an incidence matrix tying links with paths. In such models, either or is unknown. Tomography approaches infer the unknown parameters from data. The conditions under which network parameters are identifiable under sufficient data are often of interest, so that approaches can be developed that allow parameter identification. For example, Tebaldi and West (1998) considered the problem of inferring road traffic between nodes using link measurements and employed a Bayesian approach to rectify identifiability issues. Network tomography infers the path-level parameters in for example in Chen et al. (2010), or the presence of links in for example in Ni et al. (2010).