Wavelet Based Path Loss Modeling for Global System for Mobile Communication in an Urban Environment

In this paper, we measured GSM signal strength in the Dnepropetrovsk city to predict path loss in study area using one-dimensional multilevel wavelet and used neural network clustering to determine average GSM signal strength received at the study area. The wavelet predicted that the GSM signal fades 5 times the propagation frequency and attenuated with RMSE of 3.428dB; this attenuation value is used to modify the reference model. The neural network clustering revealed that -75dB is received more frequent followed by -85dB. This means that the signal strength received at the study is mostly weak signal.