Direction estimation using conjugate cyclic cross-correlation: more signals than sensors

We consider the problem of estimating the directions of arrival of multiple communication signals arriving at an uniform linear array. By considering the conjugate cyclic crosscorrelations of the sensor outputsand using a Bayesian framework, we propose a direction finding algorithm that allows us to estimate the directions of arrival for a more signals than sensors scenario. The algorithm does not need any training sequence and only requires a priori knowledge of the cyclic frequencies and the number of sources. It is also possible to estimate the directions of arrival in a multi-path fading environment under certain conditions.