DEVELOPMENT OF A MODIFIED EMPIRICAL MODEL FOR NEW MOBILE RADIO NETWORK AT 28 GHZ MILLIMETER WAVE SPECTRUM

Akinyinka Olukunle Akande; Olusola Kunle Akinde; Chukwuemeka Etus; Oluwadara Joel Odeyinka

Akinyinka Olukunle Akande Department of Electrical and Electronic Engineering, Federal University of Technology, Owerri
Olusola Kunle Akinde Department of Electrical and Biomedical Engineering, First Technical University, Ibadan, Nigeria
Chukwuemeka Etus Department of Electrical and Electronic Engineering, Federal University of Technology, Owerri, Nigeria
Oluwadara Joel Odeyinka Department of Information Management Technology, Federal University of Technology, Owerri, Nigeria

Keywords: APSO, Okumura-Hata, Modified Model, Path loss, RMSE, ECC-33, Auto-regressive

Abstract

Quality of received signal depends partly on the degree of loss the signal experiences along its path. Different models are available for network path loss (PL) prediction. The issues of poor network in some urban area need special attention especially during the deployment of 5G new radio network. It is believed that 4G and 5G networks will coexist for a very long time. 5G network will work in conjunction with the existing 4G network, and 4Gnetwork is not going away any time soon. Hence, it is important to develop a suitable model to mitigate the outcome of signal attenuation and signal interference in 5G network. This paper presented a modified particle swarm algorithm model in 4G network at 28 GHz millimeter wave spectrum. Some existing PL models are unsuitable for PL prediction in certain environments necessitating the development of an appropriate model. To achieve this, existing models and measured data were compared to find out the closest model to the measured value. The environmental loss exponent was obtained to modify Okumura-Hata which is the closest model to the measured path loss (PL) data. A new modified model for the terrain was enhanced by the developed Autoregressive Particle Swarm Optimization (APSO) Algorithm. Root Mean Square Error (RMSE) results obtained are 4.499 dB, 1.050 dB and 0.872 dB for Okumura-Hata, Modified and Enhanced model, respectively. The corresponding values for Mean Absolute Percentage Error (MAPE) are 0.254 %, 0.058 % and 0.048 %. The result showed that Enhanced APSO is a suitable model for predicting 4G network in Abuja, Nigeria