Abstract
Conference Title: 2016 International Wireless Communications and Mobile Computing Conference (IWCMC) Conference Start Date: 2016, Sept. 5 Conference End Date: 2016, Sept. 9 Conference Location: Paphos, Cyprus For a long time, developing empirical propagation models have been an area of interest of scientist and researchers due to their simplicity and low cost in term of used resources. Their major drawbacks reside in the lack of accuracy which can be improved by adjusting model for a given area and frequency. This operation is called model tuning or calibration. In the present work we are interested to present and compare two methods for tuning Standard Propagation Model (SPM) parameters to improve its accuracy in the most used bands which are 900MHz, 1800MHz, 2100MHz and 2600MHz bands. The first method is based on multi linear regression method while the second one is based on the use of neural network. The predictions of tuned models are compared to real measurements to highlight the accuracy of each one. To accomplish this work continuous wave (CW) measurements campaign is conducted in Tunis City in the four cited bands for urban, suburban and rural environments.