On the Cramer-Rao bound for direction finding of correlated signals

In this correspondence we present compact closed form formulas for the Fisher information matrix (FIM) corresponding to the joint estimation of the directions of arrival (DOA's), the signal covariance matrix, and the noise variance. Using these formulas we investigate the effect of signal correlation on the achievable accuracy of direction finding systems in a correlated signal environment. As expected, estimation accuracy decreases with increasing correlation magnitude. We observe that under certain conditions (small aperture and high correlation magnitude), the correlation phase has a strong effect on DOA estimation accuracy.