Field strength prediction in indoor environments with neural networks

Author

Wölfle, G. ; Landstorfer, F.M.

Author_Institution

Inst. fur Hochfrequenztech., Stuttgart Univ., Germany

Volume

1

fYear

1997

fDate

4-7 May 1997

Firstpage

82

Abstract

A new model for the field strength prediction for mobile communication networks inside buildings is presented. The model is based on artificial neural networks, trained with measurements. In contrast to other neural prediction models a good generalization is achieved, so the prediction results are also very accurate in buildings not used for the training of the neural network. Two algorithms for the selection of the training patterns for the neural networks are presented and compared to each other

Keywords

backpropagation; feedforward neural nets; field strength measurement; indoor radio; land mobile radio; multilayer perceptrons; radio networks; radiowave propagation; telecommunication computing; artificial neural networks; backpropagation algorithm; buildings; field strength prediction; indoor environments; measurements; mobile communication networks; multilayered feedforward perceptron; neural networks; neural prediction models; radiowave propagation; training patterns; Artificial neural networks; Buildings; Electromagnetic propagation; Electronic mail; Indoor environments; Intelligent networks; Neural networks; Predictive models; Reflection; Transmitters;

fLanguage

English

Publisher

ieee

Conference_Titel

Vehicular Technology Conference, 1997, IEEE 47th

Conference_Location

Phoenix, AZ

ISSN

1090-3038

Print_ISBN

0-7803-3659-3

Type

conf

DOI

10.1109/VETEC.1997.596323

Filename

596323