A study of LTE network performance based on data analytics and statistical modeling

Sanqing Hu, Ye Ouyang, Yu Dong Yao, M. Hosein Fallah, Wenyuan Lu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

17 Scopus citations

Abstract

Various simulation based approaches are developed to study wireless network performance and capacity traditionally. However, these simulation based approaches has its own limitations in supervised and unsupervised learning. Since big data techniques become more available, using big data techniques to understood the huge amount of the LTE network measurements and diagnosis data, evaluate and predict the LTE network capacity and performance becomes a very promising approach. This paper investigates the LTE network performance from the aspect of data analytics and statistical modeling. In this paper, we develop a methodology of data analytics and modeling to evaluate LTE network performance based on traffic measurements and service growth trends. A relational algorithm is developed to obtain the relationship between LTE network resources and LTE KPIs and a forecasting algorithm is developed to trend the network resource consumptions based on traffic and service growth. The numerical results show a high accuracy, robustness, and reliability of the proposed methodology.

Original languageEnglish
Title of host publication2014 23rd Wireless and Optical Communication Conference, WOCC 2014
DOIs
StatePublished - 2014
Event2014 23rd Wireless and Optical Communication Conference, WOCC 2014 - Newark, NJ, United States
Duration: 9 May 201410 May 2014

Publication series

Name2014 23rd Wireless and Optical Communication Conference, WOCC 2014

Conference

Conference2014 23rd Wireless and Optical Communication Conference, WOCC 2014
Country/TerritoryUnited States
CityNewark, NJ
Period9/05/1410/05/14

Keywords

  • Big Data Analytics
  • LTE
  • Network Capacity
  • Network Performance
  • Network Resource
  • Statistical Modeling

Fingerprint

Dive into the research topics of 'A study of LTE network performance based on data analytics and statistical modeling'. Together they form a unique fingerprint.

Cite this