Path loss predictions for multi-transmitter radio propagation in VHF bands using Adaptive Neuro-Fuzzy Inference System

Path loss prediction is an important process in radio network planning and optimization because it helps to understand the behaviour of radio waves in a specified propagation environment. Although several models are currently available for path loss predictions, the adoption of these models requires...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Engineering science and technology, an international journal an international journal, 2018-08, Vol.21 (4), p.679-691
Hauptverfasser: Surajudeen-Bakinde, Nazmat T., Faruk, Nasir, Popoola, Segun I., Salman, Muhammed A., Oloyede, Abdulkarim A., Olawoyin, Lukman A, Calafate, Carlos T.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Path loss prediction is an important process in radio network planning and optimization because it helps to understand the behaviour of radio waves in a specified propagation environment. Although several models are currently available for path loss predictions, the adoption of these models requires a trade-off between simplicity and accuracy. In this paper, a new path loss prediction model is developed based on an Adaptive Neuro-Fuzzy Inference System (ANFIS) for multi-transmitter radio propagation scenarios and applicable to the Very High Frequency (VHF) bands. Field measurements are performed along three driving routes used for testing within the urban environment in Ilorin, Kwara State, Nigeria, to obtain the strength values of radio signals received from three different transmitters. The transmitters propagate radio wave signals at 89.3 MHz, 103.5 MHz, and 203.25 MHz, respectively. A simple five-layer optimized ANFIS network structure is trained based on the back-propagation gradient descent algorithm so that given values of input variables (distance and frequency) are correctly mapped to corresponding path loss values. The adoption of the Pi membership function ensures better stability and faster convergence at minimum epoch. The developed ANFIS-based path loss model produced a low prediction error with Root Mean Square Error (RMSE), Standard Deviation Error (SDE), and correlation coefficient (R) values of 4.45 dB, 4.47 dB, and 0.92 respectively. When the ANFIS-based model was deployed for path loss predictions in a different but similar propagation scenario, it demonstrated a good generalization ability with RMSE, SDE, and R values of 4.46 dB, 4.49 dB, and 0.91, respectively. In conclusion, the proposed ANFIS-based path loss model offers desirable advantages in terms of simplicity, high prediction accuracy, and good generalization ability, all of them critical features for radio coverage estimation and interference feasibility studies during multi-transmitter radio network planning in the VHF bands.
ISSN:2215-0986
2215-0986
DOI:10.1016/j.jestch.2018.05.013