Efficient performance monitoring for ubiquitous virtual networks based on matrix completion

Wang, Xinheng, Xu, Chuan, Zhao, Guofeng, Xie, Kun and Yu, Shui 2018, Efficient performance monitoring for ubiquitous virtual networks based on matrix completion, IEEE access, vol. 6, pp. 14524-14536, doi: 10.1109/ACCESS.2018.2815548.

Attached Files
Name Description MIMEType Size Downloads

Title Efficient performance monitoring for ubiquitous virtual networks based on matrix completion
Author(s) Wang, Xinheng
Xu, Chuan
Zhao, Guofeng
Xie, Kun
Yu, ShuiORCID iD for Yu, Shui orcid.org/0000-0003-4485-6743
Journal name IEEE access
Volume number 6
Start page 14524
End page 14536
Total pages 13
Publisher Institute of Electrical and Electronics Engineers
Place of publication Piscataway, N.J.
Publication date 2018-03-13
ISSN 2169-3536
Keyword(s) performance monitoring
matrix completion
science & technology
computer science, information systems
engineering, electrical & electronic
computer science
wireless sensor networks
Summary Inspired by the concept of software-defined network and network function virtualization, vast virtual networks are generated to isolate and share wireless resources for different network operators. To achieve fine-grained resource control and scheduling among virtual networks (VNs), network performance monitoring is essential. However, due to limitation of hardware, real-time performance monitoring is impossible for a complete virtual network. In this paper, taking advantage of the low-rank characteristic of 90 virtual access points (VAPs) measurement data, we propose an intelligent measurement scheme, namely, adaptive and sequential sampling based on matrix completion (MC), which exploits from the MC to construct the complete data of VN performance from a partial direct monitoring data. First, to construct the initial measurement matrix, we propose a sampling correction model based on dispersion and coverage. Second, a stopping condition for the sequential sampling is introduced, based on the stopping condition, the sampling process for a period can stop without waiting for the matrix reconstruction to reach certain of accuracy level. Finally, the sampled VAPs are determined by referring the back-forth completed matrixes' normalized mean absolute error. The experiments show that our approach can achieve a constant network perception and maintain a relatively low error rate with a small sampling rate.
Language eng
DOI 10.1109/ACCESS.2018.2815548
HERDC Research category C1 Refereed article in a scholarly journal
ERA Research output type C Journal article
Copyright notice ©2018, IEEE
Persistent URL http://hdl.handle.net/10536/DRO/DU:30109525

Document type: Journal Article
Collections: School of Information Technology
Open Access Checking
Connect to link resolver
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in TR Web of Science
Scopus Citation Count Cited 0 times in Scopus
Google Scholar Search Google Scholar
Access Statistics: 6 Abstract Views, 2 File Downloads  -  Detailed Statistics
Created: Mon, 18 Jun 2018, 11:25:56 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.