Sequential adaptive sampling based on cubic spline interpolation to accelerate the near field scanning

Abstract

This paper describes a new automatic and adaptive sampling strategy to characterize radiated electromagnetic emissions. It is based on an optimized algorithm to accelerate near field scanning of printed and/or integrated circuits, electronic and/or power electronic systems. The algorithm sequentially defines where experimental measurements should be performed and which point must be captured. This approach is original; it combines both a deterministic mesh and a sequential scanning method enhanced by a cubic spline-based triangulation interpolation. This strategy seeks to decrease the time required to perform the electromagnetic compatibility (EMC) measurement without increasing the errors. This approach is applied to simulated and measured examples, which validates its effectiveness compared to the uniform grid sampling.