A measurement-based methodology for the determination of validity domains of prediction models in urban environment

The choice of a given propagation model has a major impact on the accuracy and effectiveness of cellular planning, especially in urban environment, where a higher degree of detail is required. Rigorous models, such as those based on ray optics, are generally heavily time-consuming; furthermore, they are accurate only if high-resolution databases (of buildings, land usage, etc.) are available, and the cost of these databases is increasingly affecting the operators' budget. Therefore, one should use them only in their "validity domain," i.e., the area where they are both reliable and necessary: using complex tools where it's not needed implies an overhead in terms of computation time and, above all, database cost. For this reason, increasing interest and effort are being devoted by operators, manufacturers, and system planners in general to determine the validity domains of propagation models in urban environment. In this paper, we introduce a new methodology, based on objective indicators related to field strength statistics, to determine these domains.