Wiener filter processing for directions of arrival estimation using a moving window operator

Abstract

The authors consider the estimation of the directions of arrival in a multiple-source environment using a linear array of m sensors with d sources assumed to be in the far field and generating signals. A technique based on the moving window operator was proposed to estimate the locations of these sources. A Wiener filter preprocessing scheme to improve the estimates at low signal-to-noise ratios is described. On the basis of the knowledge of second-order statistics and the assumptions of stationary and time-invariant signals, the coefficients of the Wiener filter can be easily found. It is shown that the estimates of the directions of arrival are improved at low signal-to-noise ratios. Results of computer simulations carried out to evaluate the approach are also presented.<>