High Resolution Spectrum Estimation of FMCW Radar Signals

Abstract

This paper presents a comparative study of high resolution spectral estimation methods applied to Frequency Modulated Continuous Wave (FMCW) radar data; autoregressive (AR), Prony, and, Eigenanalysis-based methods are briefly reviewed. A comparison of spectral lines detection capabilities between methods for different signal to noise ratio (SNR) and modal amplitudes is made using numerical experiments. Computer simulations have been made using a test signal with ten frequencies in order to evaluate the probability of detection of each frequency and the bias. High resolution spectral estimates from real data using MUSIC 141 and the modified covariance method [2] (AR) showed much more spectral details than the Fourier Transform approach, and it exhibited consistent peak position compared to the lower resolution Fourier spectrum. This work was supported by the DREV under contrat no. W7701-22440/01 -XSK.