This research article published by Computer Engineering and Intelligent Systems, Vol.5, No.9, 2014
Empirical propagation models have found favor in both research and industrial communities owing to their speed
of execution and their limited reliance on detailed knowledge of the terrain. In mobile communication the
accuracy prediction of path losses is a crucial element during network planning and optimization. However, the
existence of multiple propagation models means that there is no propagation model which is precisely and
accurate in prediction of path loss fit for every environs other than in which they were designed. This paper
presents few empirical models suitable for path loss prediction in mobile communication. Experimental
measurements of received power for the 900 MHz GSM system are made in urban, suburban, and rural areas of
Dar es Salaam, Tanzania. Measured data are compared with those obtained by five prediction models: Stanford
University Interim (SUI) models [1], the COST-231 Hata model [2], the ECC-33 model [3], the ERICSSON
model [4], and the HATA-OKUMURA model [5]. The results show that in general the SUI, COST-231,
ERICSSON, and Hata-Okumura under-predict the path loss in all environments, while the ECC-33 model shows
the best results, especially in suburban and over-predict pathloss in urban area.