Improving RCS and ISAR image prediction of terrestrial targets using random surface texture

Tools for electromagnetic simulation, such as SE-RAY-EM/FERMAT, make RCS and ISAR image computations down to millimetre wavelength possible from the CAD of targets. The higher the frequency is, the greater the influence of the precision and the realism of the CAD models on results are. In millimetre wavelength, whatever the precision of the CAD, the roughness of surface and the non flatness of sheets of construction of terrestrial military vehicles cannot be reproduced by the geometry. Nevertheless, a realistic prediction of the RCS can be carried out by using statistical random surface superimposed on the deterministic initial geometry of the CAD. The calculations carried out on the 80 000 and 600 000 facetted CAD of the ZSU 23-4 illustrate the efficiency of this process.