Open AccessEcho state network‐based radio signal strength prediction for wireless communication in Northern Namibia
Author(s) -
Gideon Kenneth,
Nyirenda Clement,
TemanehNyah Clement
Publication year - 2017
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.2016.1290
Subject(s) - echo state network , python (programming language) , computer science , goodness of fit , computation , echo (communications protocol) , wireless , support vector machine , signal strength , radio propagation , radio signal , radio frequency , algorithm , data mining , artificial intelligence , machine learning , artificial neural network , telecommunications , computer network , recurrent neural network , operating system
A method for predicting radio signal strength using echo state networks (ESNs) is proposed and applied to three different locations in Northern Namibia. This method aims at providing a better way for radio signal strength prediction, which leads to improvements in wireless communication planning, design and analysis. Its performance is compared with the support vector regression (SVR) method optimised for radio propagation modelling. Simulations are conducted in Python using propagation data measured from the three locations based on the following four performance measures: goodness of fit criteria; error measures; computation complexities; and F ‐test for statistical model comparison. Simulation results show that the ESN method gives a better prediction accuracy in terms of the goodness of fit criteria and the error measures; however, it is inferior to the SVR method in terms of computation complexities. In addition, results from the F ‐test also indicate that the ESN method provides a significantly better fit than the SVR method.