In wireless communication, signal is transmitted by a transmitting antenna and received by a receiving antenna, hence, any distortion in signal strength at the receiving end is known as path loss. All transmitted information/signal or electromagnetic waves incur path loss from source to destination due to reflection, refraction, diffraction, multipath, interference and scattering. Path loss is also influenced by geographical terrain contours, environment and propagation medium. The distance between the transmitter and the receiver, and the height of antennas. It is therefore intractable to formulate the exact path loss model for all locations. One of the underlying difficulties with the application of a predicting path loss model for any environment is that no two areas are identical in the composition of buildings geometry, terrains, and atmospheric conditions among other visible parameters. In order to overcome some of the problems stated above, parameters of certain propagation models must be evaluated and optimized with reference to a target environment. Thus, accurate path loss model is useful to improve the quality of service, reduce undesirable power losses, increase coverage area, and determine the best arrangement of base stations as the case may be. This research project has presented a comparative study between measurement surveys on live signal transmission of GSM service providers situated in some selected states in South-West Nigeria (Ondo state, Osun state, Ekiti State). In order to examine the goodness of logarithmic fit of existing pathloss model to the field data, the Root Mean Square Errors (RMSE) were also evaluated. Incorporating the effect of top view of buildings on the diffraction of rooftop fields down to street level is another important strategy used for ensuring optimized performance of mobile cellular networks. In this thesis, several cases of top view of building constructions were examined. It was observed that the nature of the rooftop view of buildings resulted in deviations of sector average signal which ranged from 1.2 to 4.72 dB. From Table 4.1b and 4.1c the Walfisch Betoni model outperformed the other models with RMSE of 6.36 – 8.67 for Ondo State, 6.34 – 7.18 for Osun State and 6.27 – 6.99 for Ekiti State. Based on the closest agreement of field data, the Walfisch Betoni model is the most suitable propagation model in these States. The overall results will serve as good tools to engineers and scientist directly involved in cellular network planning of the South-West geopolitical zones of Nigeria.