A new complexity reduced direction-of-arrival estimation method for highly correlated wave fronts

In this paper, we present a complexity reduced version of the DEESE algorithm proposed by D. Grenier (1996) for direction-of-arrival estimation of highly correlated wave fronts with an uniform linear antenna array. The DEESE algorithm resides in applying a preprocessing to the estimated spatial correlation matrix before using the MUSIC method, which implies an eigenvalue decomposition of the correlation matrix. We will show that such a preprocessing can be applied directly to the sample vectors. This allows to significantly reduce the dimension of the eigenvalue problem and hence, to significantly reduce the execution time of the algorithm. Despite its reduced complexity, our algorithm gives better performances than the common spatial smoothing method.