ARBITRARY VOXEL SELECTION FOR ACCELERATING A RAY TRACING-BASED FIELD PREDICTION MODEL IN URBAN ENVIRONMENTS

Abstract

Great accuracy and reasonable computational time are desirable in a deterministic ray tracing model for an e-cient radio frequency planning. An algorithm to speed up a ray tracing engine is described which allows to select arbitrary areas around transmitters and receivers by dividing the scene in a voxel chessboard. The reliability of the algorithm has been evaluated by comparing measured and predicted path losses in real urban scenarios while the algorithm performance is expressed in terms of computational time reduction. The goal of propagation modeling is to describe with a mathematical formulation the spreading of electromagnetic waves inside the considered scenario. Many empirical, statistical and physical models are used for this purpose. Among the asymptotic methods, inverse 3D deterministic Ray Tracing (RT) models are characterized by an accurate prediction of the electromagnetic fleld both in the case of an outdoor site as well as an indoor location. In Radio Frequency (RF) system planning, the RT propagation model and data of the scenario can be used to predict the RF coverage area of a transmitter (Tx). This analysis allows to determine the received signal strength at the receiver (Rx), the Path Loss (PL) of a wireless link and channel impairment such as delay spread due to multi-path fading. The knowledge of these parameters before the wireless system setting- up leads to an installation cost reduction together with a quality of services increase. For example, it is possible to estimate the minimum number of antennas and their best locations in order to guarantee a