Unitary Cyclic MUSIC for Direction Finding in GPS Receivers

Several anti-jam techniques for GPS receivers rely on the direction-of-arrival (DOA) of the satellite signals. These angles change with moving GPS receiver platforms and, as such, they should be constantly estimated for the purpose of jammer nulling and beamforming. In this paper, we propose a new approach for directional finding for multi-antenna GPS receivers. This approach is based on a new unitary-cyclic subspace method, which utilizes the modulation and repetition properties of the GPS code to discriminate against jammer signals. We apply the forward-backward spatial smoothing technique to the cyclic array covariance matrix to decorrelate the direct path from multipath, so nulling of the undesired signals can occur. A unitary formulation is considered for real-valued eigendecomposition which aims at reducing the payload in the GPS receivers. Simulations are provided showing that the proposed approach has better performance than existing high-resolution methods that neither apply cyclostationary or unitary transformations for DOA estimation