Statistical Analysis and Modeling of a Quadrotor’s Radar Cross-Section

Resumo

This paper focuses on the statistical analysis and modeling of the radar cross-section (RCS) of a DJI Phantom IV drone. The RCS datasets are generated by means of simulations performed at 9.41 GHz for distinct azimuth and elevation angles. Further, these datasets are fitted to usual probability distributions using three criteria, namely log-likelihood (LLK), Akaike information criterion (AIC), and Bayesian information criterion (BIC). Moreover, the impacts of RCS modeling on the radar detection range is analyzed. Based on numerical results, the Exponential distribution is shown to be the best fit for the RCS datasets. A good agreement is obtained between the Exponential’s probability distri- bution function and the histogram of datasets. Finally, the use of this distribution for modeling RCS datasets achieves a maximum error of 3.8% when applied to the radar range equation.