ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

A Generalized Regression Neural Network Approach to Wireless Signal Strength Prediction

Journal: International Journal of Trend in Scientific Research and Development (Vol.4, No. 3)

Publication Date:

Authors : ;

Page : 467-471

Keywords : Artificial Intelligence; Field Strength; Generalized Regression Neural Network; Okumura Model; COST 231 Hata Model;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

This study presents a Generalized Regression Neural network GRNN based approach to wireless communication network field strength prediction. As case study, the rural area between the cities of Bauchi and Gombe, Nigeria, was considered. The GRNN based predictor was created, validated and tested with field strength data recorded from multiple Base Transceiver Stations at a frequency of 1800MHz. Results indicate that the GRNN based model with Root Mean Squared Error RMSE value of 5.8dBm offers significant improvements over the empirical Okumura and COST 231 Hata models. While the Okumura model overestimates the field strength, the COST 231 Hata significantly underestimates it. Finangwai D. Jacob | Deme C. Abraham | Gurumdimma Y. Nentawe "A Generalized Regression Neural Network Approach to Wireless Signal Strength Prediction" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-4 | Issue-3 , April 2020, URL: https://www.ijtsrd.com/papers/ijtsrd30501.pdf

Last modified: 2020-06-09 14:57:09