Arbitrary voxel selection for speeding up a ray tracing-based EM simulator

Over the last decades many empirical, statistical and physical models for channel modeling have been proposed. Among the asymptotic methods, inverse 3D deterministic Ray Tracing (RT) models are characterized by an accurate prediction of the electromagnetic field both in the case of an outdoor as well as an indoor scenario. For Radio Frequency (RF) system planning, the RF coverage analysis uses the RT propagation model and the data of the scenario to predict the e. m. field distribution radiated by a transmitter (Tx). Therefore it could be necessary to determine the signal strength at the receiver (Rx), the Path Loss (PL) of a wireless link or channel impairment such as delay spread due to multi-path fading. The time required by the RT algorithm to compute all rays connecting Txs and Rxs exponentially increases as the number of objects in the scenario linearly increases. Hence, this model is often considered time expensive if compared to the empirical and statistical propagation ones. Recently, a simplification of the scenario based on a heuristic pre-processing has been proposed [1] to reduce the computational time. This technique, given the Tx and Rx positions, identifies the active part of the scene database selecting all buildings located within an ellipse of focuses Tx and Rx. However, when the distance between Tx and Rx and the largest scene dimension are comparable in order of magnitude, the ellipse surface almost overlies the whole scenario and the method becomes inefficient. To overcome this drawback due to an intrinsic geometric limit, we propose a preprocessing refinement of the scene characterized by a higher degree of flexibility of the selected area that will be subsequently analysed by the RT method. By using a grid space division, we can select all cells crossed by the line connecting Tx and Rx and/or the nearest ones. Moreover, if desired, it is also possible for the designer to select only the voxels crossed by the street or district under analysis.