An Optimization Approach for Predicting Worst-Case Positions in EMI Final Measurement Based on Convolution Neural Network

In this paper, we present an improvement in existing Electromagnetic Interference (EMI) measurement according to the norm FCC§ 15.209 in the range 30MHz to 1GHz. A developed measurement tool and a convolution neural network(CNN) were used to reduce the required time to carry out the final measurement on critical frequencies by predicting the radiation emission and then determining the position azimuth of the turntable and the height of the antenna that meet the maximum radiated emission level. The neural network was trained using real EMI measurements which were performed in the Semi Anechoic Chamber(SAC) by Cetecom GmbH in Essen, Germany.