|Radio propagation models characterize the radio wave propagation on the basis of radio propagation mechanisms and of BS antennas positioning and electrical parameters. By the aim of these models the range of a wireless communication system can be calculated or predicted.
Different models are available but each of them calculates the Path Loss L between the transmitting and receiving antennas in a wireless communication systems. The Path Loss in a wireless communication system (as shown in figure) is the ratio of the transmitted power to the received power, usually expressed in decibels. It includes all of the possible elements of loss given by the interaction mechanisms between the propagating wave and any object between the transmitting and receiving antennas.
These radio propagation models can be classified considering different aspects regarding not only the wireless system considered but also the objects between transmitters and receivers, the operating frequency and so on.
A common classification describes radio propagation models in such a way:
- ideal models: only the distance between the transmitters and receivers influence the Path Loss calculation. They can be applied only in ideal cases rather than in real ones. (i.e.: Free Space Loss model)
- empirical models: created by making an extensive set of actual path loss measurements in a given geographical area and by fitting to these measurements an appropriate function the parameters of which are derived for the specific environment, frequency and antenna heights considered, in order to minimize the error between the model and the measurements. Such models can be used to design systems operating in similar environments to the original ones. (i.e.: COST231-Okumura-Hata). Empirical models can be implemented rapidly without requiring any extremely accurate or expensive geographical databases. These models are robust since they are as a general
- semi-empirical models (or semi-deterministic models): they combine the analytical formulation of physical phenomena like reflection, transmission, diffraction and scattering with a statistical fitting by variables adjustment using experimental measurements. These methods are more robust than purely empirical models since they avoid the improbability of independent variables (Ex: COST231-Walfisch-Ikegami)
- deterministic models (or theoretical models): they are based on the fundamental laws of physics combined with adequate approximations and with environment (atmosphere and land) models. These models rely on complex mathematical relationships and require the resolution of Maxwell’s equations through the use of different analysis methods: finite difference and finite element methods, parabolic equation method, physical and geometrical optics methods. They require relatively high computation times but they are very accurate and site-specific. (Ex: Ray Launching).
For example, ray models are deterministic models relying on an accurate knowledge of the actual propagation environment and therefore requiring the use of geographical databases. These models can be used for the prediction of the different propagation paths (rays) in a given environment