Parameter Estimation from Noisy Transient Electromagnetic Measurements

摘要

The study of the EMP phenomenon has promoted the development of techniques to investigate transient electromagnetic response data. Characterization of the EMP transient response information is necessary to evaluate the performance of that system in a hostile environment. An efficient technique to characterize this performance is to fit an electromagnetic model to the data. The performance of three different signal processing techniques applied to parameterize a body from noisy experimental electromagnetic transient response data is described. The techniques, which range from the well-known Prony method to the more sophisticated extended Kalman filter and finally to the highly sophisticated maximum likelihood identifier, are briefly described. The performance of these algorithms is compared, and their tradeoffs are discussed. 14 figures, 2 tables. (ERA citation 05:035786)