Optimization of EMC aerospace margins using re-sampling techniques with Monte Carlo simulations

This contribution aims to illustrate the interest of re-sampling technique for accurate and efficient assessment of confidence intervals. Taking exhaustively into account uncertain inputs of complex systems theoretically requires to model an infinite number of configurations (i.e. to properly compute probability density function of random output). Sampling techniques such as Monte Carlo (MC) simulations are reference methods but are also restricted due to their slow convergence rate. This study offers a computational framework to validate the use of re-sampling techniques for an aerospace EMC immunity issue. A particular attention will be given to the assessment of confidence intervals of currents due to field-to-wire coupling in reverberating environments. The robustness and accuracy of the method will be validated by considering high-orders statistics and quantiles based upon time domain MC simulations.