Requirements for Indoor Building Databases to increase the Accuracy of the Propagation Results

Currently, for the planning of wireless networks (e.g. WLAN) in indoor scenarios either empirical (direct ray), semi-deterministic or ray-optical propagation models are used. In this paper two different model approaches are used to investigate the impact of database inaccuracies to prediction results. Therefore a fast ray tracing model using a preprocessing of building vector data is used, as well as a semi-deterministic dominant path model. This paper focuses on different inaccuracies in indoor building databases: Time variance due to doors or windows which are closed or opened causes slow fading and has thus significant influence to the prediction results. The accuracy of models is also influenced by missing objects in the building database, such as furniture or walls. The impact of adjacent buildings is also investigated.