Abstract
Network topology is important information for many network control and management applications. Network tomography infers network topology from end-to-end measured packet delays or losses, which is more feasible than internal cooperation-based methods and attracts many studies. Most of the existing methods for network topology inference usually function under the assumption that the distribution of packet delay or loss follows a given distribution (e.g., Gaussian or Gaussian mixture), and they estimate network topology from the parameters of the given distribution. However, these methods may fail to obtain an accurate estimation because the real distribution of packet delay or loss usually cannot be described by a certain distribution. In this paper, we present a novel network topology inference method based on the unicast end-to-end measured delays. The method abandons the assumption of packet delay distribution and constructs network topology by inferring the higher-order cumulants of internal links from the end-to-end measured delays. The analytical and simulation results show that the proposed method offers over 10% improvement in accuracy compared with that of the state-of-the-art works.
Highlights
As size of the Internet has grown dramatically, it has become a giant system with a complex structure
The existing methods for network topology measurement can be divided into two categories: internal cooperation-based methods and network tomography-based methods
The network tomography-based methods are more feasible than the internal cooperation-based methods because they are capable of obtaining the topology without the cooperation of internal nodes
Summary
As size of the Internet has grown dramatically, it has become a giant system with a complex structure. The complexity of traffic distribution (such as traffic imbalance and nonstationary nature) in the actual network makes a fixed threshold ineffective, and such a threshold may not be suitable for all links and nodes To solve this problem, this paper first designs a link walking algorithm to cluster the links in the binary tree and uses a two-state automaton model with Bayes inference to identify the true or false links (the links can be deleted). Compared with the existing methods, the proposed approach is capable of obtaining more accurate topology estimation because we use the delay cumulants of second order and above to infer the topology, and the statistical information of the path delays can be more fully utilized in our method.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
