Abstract:

In order to improve the accuracy of direction finding of distributed non-identical subarray antenna radar which cannot apply the estimation of signal parameters via the rotation invariant technique algorithm (ESPRIT), a new method based on root rank reduction estimator (root-RARE) is proposed for estimating the direction of arrival (DOA) of the target. The length of the distributed subarray antenna baselines is longer than the half wavelength, which results in grating lobes in the combined array beampattern and ambiguity in DOA estimation. An estimation of signal parameters via the combination of the root-RARE and multiple signal classification (MUSIC) is used to resolve the ambiguities. The root-RARE yields unambiguous but highvariance direction estimates to disambiguate lowvariance but cyclically ambiguous estimates from the MUSIC algorithm with the whole distributed subarray antennas, then lower variance and non-ambiguous estimates are achieved. The Cramer-Rao bound of direction is calculated. The performance of DOA estimation and the relation between the baseline ambiguity threshold and the signal to noise ratio threshold are analyzed. Simulation results demonstrate the validity and effectiveness of the proposed algorithm.