Analysis of path loss models for millimeter wave system n urban cellular environment

Abstract

This research is to identify the parameters of two models of path loss focusing on the frequency band that has high potential to be used for 5G carrier frequency. Those two path loss models are Close-In (CI) Reference Model and Alpha-Beta-Gamma (ABG) Model. The potential frequency band is band 22 that generally known to operate at 3.5 GHz. The loss characteristics of the frequency band is very crucial to be understood as it will be given a huge impact in many important things in Radio Access Network (RAN), such as in planning level, signal threshold, hand over to neighbouring cell, noise level, coverage distance, Efficient Isotropic Radiated Power (EIRP) range, Ec/Io, co-channel interference, adjacent and channel interference. When the parameters are known, the sufficient signal power level to enable mobile users to stay connected within the coverage area can be precisely predicted. The environment of research being conducted are outdoor and indoor for Line Of Sight (LOS) urban condition. Both uplink and downlink of band 22 were transmitted and the signal power at certain distances have been measured to determine the signal path loss. By using the Matlab GUI, all parameters were successfully obtained. The results show that the parameters obtained are stables for both path loss models. For CI model, the parameter of Path Loss Exponent, n is within 1.763<n<3.388, where for ABG model, the results of parameters ranges are 0.321<?<16.989, 176.533<?<180.068 and -43.832<?<26.604. In terms of consistency, the CI model shows consistence parameter compared to ABG model because it is only involve one parameter, n, but in terms of accuracy, ABG model is more accurate due to the complexity of equation and the existing of the floating parameter. One of the limitation of this research is the environment that the test being conducted was only in LOS condition, thus, in the Non-Line Of Sight environment, the parameters might be different. In addition, the low EIRP of the transmitter for this research also has affected the measurements distances of the signal strength.