Development of an empirical power model and path loss investigations for dense urban region in Southern India

The main factor in cellular network planning is the accurate estimation of the received signal strength with a deeper understanding of the characterisation of a radio channel. A detailed planning is required which concentrates on parameter estimation with necessary field measurements. In this paper a drive test is conducted in the dense urban region of Hyderabad city (Southern India), using Test Mobile System (TEMS) tool to record the received GSM 900 MHz mobile signals. Processing the measured field strengths, the path loss exponents for the measurement sites are determined using Minimum Mean Square Error method. They are found to vary from 3.35 to 4.31, which suggest that it is a shadowed urban region. An empirical power model is developed to predict the received power. Further, path loss investigations are done by implementing three standard models namely, Cost-231 Hata, Stanford University Interim (SUI) model and ECC-33 model. The performances of the models are evaluated by comparing the predictions with the measured data. From the results, it is observed that the prediction error for Cost-231 Hata model is 5.98 and its relative error is 5.10% which are least compared to the implemented models. Various statistical metrics are compared for the measured and predicted models. Overall, this paper performs power modeling, investigates the path loss over a dense urban terrain and justifies that Cost-231 Hata model has a best fit with the experimentally measured data, compared to SUI and ECC-33 model for the specified region.