Parameter estimation in chirped signals

Abstract

The discrete Fourier transform (DFT) is used almost routinely to analyze sampled time series. A careful formulation of any problem to be solved and a meticulous application of probability theory ought to generate the tools to be used for analyzing the problem. The tools thus arrived at may or may not be closely related to the DFT. In this work, the implementation of such a program is outlined in a well-known problem with its roots in radar and sonar. The transform that is germane to the analysis of a chirped signal is obtained automatically.<>