Optimization of Modified Empirical Model in 2.3 GHz Long Term Evolution Network. Case Study of FUTO

Path Loss (PL) is a factor that contributes to signal impairment in a wireless channel and can be predicted using empirical models such as Okumura-Hata, Egli, COST, and ECC-33. These models are known to be inaccurate when used in areas different from where they are meant for. In this paper an empirical model suitable for PL prediction in 4G Long Term Evolution (LTE) network was modified and optimized. Measured PL values were obtained and compared with the existing models in order to obtain the closest model to the measured data. The Okumura-Hata model as the closest to the measured values was modified using the loss exponent obtained for FUTO terrain and optimized for better performance using the developed autoregressive (AR) model. The measured data, the existing, modified and optimized AR [opt(AR m )] models were simulated using the MATLAB software. The performances of the existing and opt(AR m ) models were evaluated using the root mean square error (RMSE). The results obtained show that the proposed opt(AR m ) model is the best as compared to the existing models due to its lowest RMSE value, and thus it can be used to estimate PL for mobile radio signal in FUTO (Federal University of Technology, Owerri), Nigeria.