IEEE Transactions on Electromagnetic Compatibility · 2019 · 17 citations · 18 references
EngineeringElectrical EngineeringAntennasMicrowave EngineeringSpatial StatisticsElectromagnetic CompatibilityField TestingMicrowave PropagationAnderson-darling TestExperimental ValidationExtreme Value AnalysisExtreme Value DistributionsGeneralized Extreme Value DistributionGeneralized Extreme Value ModelsImmunity TestingNonparametric Estimation
In the framework of immunity testing, there has been recently a great interest to determine the maximum field distribution in a nested reverberation chamber, the inner cavity being considered as the equipment under test (EUT). The generalized extreme value distribution is a generalization of three asymptotic distributions that does not require knowledge of the field sample parent distribution in comparison with traditional method. It can therefore be applied in situations when the parent distribution is unknown, such as when an EUT is reverberant or is operating in the undermoded regime. In this paper, we present an experimental validation of the maximum statistical distribution of the internal field samples inside different nested cavity configurations. The Anderson-Darling test is applied to verify if a dataset is collected from a single statistical population inside the EUT with different stirring scenarios and two different EUT aperture dimensions. Parametric and nonparametric estimation are then used to model and to verify the results. The parametric and nonparametric distributions show good agreement for all test configurations considered, extending earlier works which considered reverberation-chamber fields themselves with no EUT included or did not include EUT internal stirring.
18
System identification—Theory for the user
Naresh K. Sinha · Automatica · 1989 · 9.2K citations
R.A. Fisher and the making of maximum likelihood 1912-1922
John Aldrich · Statistical Science · 1997 · 617 citations · Full text