Time domain deconvolution of transient radar data

A simple technique is presented for obtaining the impulse response of a conducting radar target by deconvolving the measured response of the experimental system from the measured response of the target. The scheme uses singular value decomposition (SVD). It is shown that the ill-conditioned nature of the deconvolution process can be linked to the use of inappropriate frequency information in the representation of the unknown impulse response. Examples using measured transient radar data establish the usefulness of the scheme. The noise sensitivity of the deconvolution process can also be traced to the improper representation of the impulse response. Accurate deconvolution is demonstrated even in the presence of large amounts of random noise.>