Impact of Building Database Accuracy on Predictions with Wave Propagation Models in Urban Scenarios

The accuracy of predictions with wave propagation models depends significantly on the quality of the building databases, as they show discrepancies when compared to reality. This is due to the inaccuracies of the photometric methods used to create these databases, but also to the fact that new buildings are raised or old buildings are pulled down. However, propagation models should feature a certain robustness against these inaccuracies. In this paper, the sensitivity of different propagation models for urban scenarios with respect to building database inaccuracies is analyzed. The main focus is the comparison of the dominant path model with a Ray Tracing model, but the results are also put in relation to the empirical COST 231 Walfisch-Ikegami model. Furthermore, the possibilities of an automatic database simplification to reduce calculation times are investigated, and it is shown that ray tracing computation times can be reduced up to 50% with only slight decreases in the accuracy by applying suitable simplification algorithms. For the evaluation of the influence on the accuracy also path loss measurements in different scenarios were used