Amplitude-only versus amplitude-phase estimation

The feasibility of transfer function estimation based on amplitude-only measurements is verified. Its properties are compared to those of a full complex estimation using both the magnitude and the phase information of the estimated transfer function. The Fisher information matrix is used as a basis of comparison. It is found that the amplitude-only estimation has a larger standard deviation on the transfer function parameters. Optimal excitation for the analyzed examples is calculated for both methods. The main difference between them is that amplitude estimates require more energy further in the transition band. The variance of the amplitude estimates is highly dependent on the excitation bandwidth used. Care has to be taken to provide sufficient excitation in the transition band(s) of the device under test to obtain sensible results with amplitude estimates.<
>