Inferring the Probability Distribution of the Electromagnetic Susceptibility of Equipment from a Limited Set of Data

Failure risk assessment of electronic equipment to an electromagnetic aggression is the cornerstone of Intentional Electromagnetic Interference (IEMI). Such a failure may occur if the electromagnetic constraint reaches a threshold that is likely to produce a dysfunction. Due to the production variability of electric / electronic equipment under analysis, its susceptibility level may be considered as a random variable. Estimation of its distribution through susceptibility measurements of a limited set of available equipment is required. We compare the performance of the Bayesian Inference (BI) and the Maximum Likelihood Inference (MLI) according to their ability to choose the true distribution for different sample size, when the true distribution is theoretically known. Then we compare the performance of the BI and the MLI on a virtual electronic device for which the true distribution is not known a priori. We finally discuss the respective benefits of BI and MLI in estimating the failure probability of equipment.