Statistical distribution of the radar cross section of various canonical targets

A numerical experiment is performed using a Shooting and Bouncing Rays based solver on eight finite clusters of eight perfectly electrically conductive canonical target geometries where each finite cluster volume includes certain amount of randomly translated and rotated target of one kind. The best probability density function that best fits to the obtained mono-static radar cross section distribution is investigated among exponential, gamma, Weibull, and log-normal distributions. Numerical results show how probability density functions for considered targets with fluctuating radar cross sections deviate from Swerling-Weibull models especially from a point of view considering the stealth measures in target designs.