Optimization of Hata Model for 2.1 GHz Communication Networks in Owerri Metropolis Nigeria

Abstract

It is often frustrating to experience call drops and poor signal reception, especially when it is motivated by an environmental
factor. This reiterates the need for adequate planning of communication signal facilities. Integrating the use of tuned models is of
great importance since it enhances the propagation of signals and improves the Quality of Service (QoS). In this work, signal
reception was enhanced through model optimization using the least square method. Drive test approach for the 3G network at 2.1 GHz in Owerri metropolis Nigeria was used to capture data from the environment. Predictions using Hata and COST 231 models were done. Hata model was chosen for optimization. Validated results show the Mean Square Error (MSE) of 31.39, 43.51 and 25.95; the Mean Absolute Percentage Error (MAPE) of 23.98%, 34.84% and 20.25% for Hata, COST 231 and optimized models
respectively. The result showed minimal error prediction for the optimized model, which implies that it can be used for signal prediction for the environment under study.