Applying the minimum free energy spectral estimation algorithm to ISAR imaging

Inverse synthetic aperture radar (ISAR) is a technique used to obtain an image of the radar target. It can be described as a spectral analysis problem. This paper discusses the application of the minimum free energy (MFE) spectral estimation algorithm for the purpose of ISAR imaging. The MFE algorithm is a regularization algorithm which achieves a better noise performance than other methods studied previously. The paper discusses the choice of parameters for the MFE technique, and presents ISAR images obtained using MFE for various observation times. MFE indeed achieves a 3-4 times higher resolution than the classic Fourier technique, without prominent spikes in the noise region. However, some of the details observed in the Fourier image are lost. Using diversity (incoherent) combining, it is shown that based upon the same RF data, MFE and the Fourier techniques produce images of similar quality, although MFE poses a less stringent requirement on the length of the coherent observation time. The MFE can be applied to ISAR imaging of close to radial targets.