A comparison of the maximum entropy method and the periodogram method applied to the spectral analysis of computer-simulated radar clutter

The maximum entropy method (MEM) for spectral analysis is described and compared to Welch's procedure using the fast Fourier transform (FFT) algorithm. Comparative results are presented for the application of these two spectral analysis techniques to computer-generated radar clutter data produced using a model developed by Hawkes and Haykin. These results show that the resolution capabilities of the MEM are superior to Welch's procedure. This feature is particularly notable in the analysis of very short clutter data records.