Open AccessGenetic algorithm‐based cellular network optimisation considering positioning applications
Author(s) -
Campos Rafael Saraiva,
Lovisolo Lisandro
Publication year - 2019
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2018.5125
Subject(s) - computer science , weighting , cellular network , latency (audio) , genetic algorithm , software deployment , wireless network , algorithm , wireless , hybrid positioning system , real time computing , computer network , machine learning , telecommunications , positioning system , mathematics , medicine , geometry , point (geometry) , radiology , operating system
This work examines the use of genetic algorithms (GAs) to optimise two different problems regarding positioning applications in wireless communication systems. The first problem involves speeding‐up the search for the mobile station (MS) position in cellular networks, when using a database correlation (DCM) technique. While addressing this problem, this work investigates three different approaches for using GA ( DcmGaFull , DcmGaBs and DcmGaBS+RTT ) and evaluates their performances (in terms of computational complexity and positioning precision) in two different regions. The performance of those methods was compared with those of the standard fall‐back methods for emergency call locating in 2G, 3G and 4G networks: cell identity (CID) and enhanced CID (ECID). Method DcmGaBs+RTT , originally proposed in this work, achieved the best performance, both in terms of lower latency and higher precision, in the two test areas. The second problem concerned optimising the deployment of a new BTS in two different test regions, while considering parameters such as coverage, traffic density and CID and ECID positioning error. This was done considering also the mixed objective of weighting the positioning error by the traffic demand, as a higher number of emergency calls are expected to be generated from areas with higher traffic density.