The Adaptive Multi-Resolution Frequency-Domain ParFlow (MR-FDPF) Method for Indoor Radio Wave Propagation Simulation. Part I : Theory and Algorithms
Résumé
This report presents the theoretical background and new developments of the multi-resolution frequency domain ParFlow (MR-FDPF) approach for the calculus or radio propagation in Indoor environments for centimetric waves. This method has been developed to face the need of a best understanding of Indoor propagation and to help the WiFi network planning task. Indeed, the development of a wireless design tool is based firstly on a radio propagation engine to predict accurately the radio coverage of access points, with a limited computational load. Usual approaches in the literature are based on either empiric modeling, deducted from measurements, or geometrical optic formalism leading to ray-tracing. While the former suffers a lake of accuracy, the later needs a trade-off between accuracy and computational load, often difficult to assess. The approach proposed herein is based on a finite element approach. Once the problem developed in the frequency domain, the linear system thus obtained is solved in two steps: a pre-processing step which consists in an adaptive multi-resolution (multi-grid) pre-conditioning and a propagation step. The second step computes the coverage of a point source with an up-and-down propagation through the binary tree associated with the multi-resolution description. This approach solves exactly the linear system but with a strongly reduced computational complexity when compared to the time domain approach. For example, a full AP coverage at a macroscopic resolution and for an environment of 1000x600 pixels (i.e. $6000m2$ at a $10cm$ resolution) lasts less than $200ms$.
Loading...