A Novel Way to Automatically Plan Cellular Networks Supported by Linear Programming and Cloud Computing

With the increasing number of mobile subscribers worldwide, there is a need for fast and reliable algorithms for planning/optimization of mobile networks, especially because, in order to maintain a network’s quality of service, an operator might need to deploy more equipment. This paper presents a q...

Full description

Saved in:
Bibliographic Details
Published inApplied sciences Vol. 10; no. 9; p. 3072
Main Authors Godinho, André, Fernandes, Daniel, Soares, Gabriela, Pina, Paulo, Sebastião, Pedro, Correia, Américo, Ferreira, Lucio S.
Format Journal Article
LanguageEnglish
Published Basel MDPI AG 01.05.2020
Subjects
Algorithms
Artificial intelligence
Cellular communications
cellular-planning
Cloud computing
cloud-services
implementation
integer linear-programming
Internet access
Linear programming
Mobile communication systems
monitoring
Optimization
Planning
Quality of service
Quality of service architectures
Software services
Technology application
Topography
Wireless networks
Portugal
Online AccessGet full text

Cover

More Information
Summary:With the increasing number of mobile subscribers worldwide, there is a need for fast and reliable algorithms for planning/optimization of mobile networks, especially because, in order to maintain a network’s quality of service, an operator might need to deploy more equipment. This paper presents a quick and reliable way to automatically plan a set of frequencies in a cellular network, using both cloud technologies and linear programming. We evaluate our pattern in a realistic scenario of a Global System for Mobile communications protocol (GSM) network and compare the results to another already implemented commercial tool. Results show that even though network quality was similar, our algorithm was twelve times faster and used four times less memory. It was also able to frequency plan seventy cells simultaneously in less than three minutes. This mechanism was successfully integrated in the professional tool Metric, and is currently being used for cellular planning. Its extension for application to 3/4/5G networks is under study.
ISSN:2076-3417
2076-3417
DOI:10.3390/app10093072