In the process of undertaking wireless signal propagation modelling, different methods have been used including deterministic and empirical models. This study is aimed at comparing the performance of predicting wireless signal propagation using Adaptive Neural Fuzzy Inference System (ANFIS), log10 distance (LOG10D)-ANFIS and LOG10D Particle Swarm Optimization (PSO) trained ANFIS for universal theoretical wireless signal prediction modelling. The last two being a modification of the original ANFIS. The predicted and target signal strength mean error (ME), root mean square error (RMSE) and standard deviation (SD) parameters were determined and compared. The study was undertaken using the one slope and two ray ground reflection models in the process of obtaining the data to be predicted. The obtained values were then used in the modeling process where it was found that the LOG10D-PSO-ANFIS model gave the closest prediction as compared to those of LOG10DANFIS and ANFIS models.
CITATION STYLE
Malack, O., Philip, K., & Peter, K. (2023). Comparing the Performance of ANFIS, LOG10-ANFIS and LOG10-PSO-ANFIS for Universal Theoretical Wireless Signal Propagation Prediction Modelling. In Lecture Notes in Electrical Engineering (Vol. 892, pp. 191–206). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-19-1645-8_20
Mendeley helps you to discover research relevant for your work.