Direction-of-Arrival Estimation for Arbitrary Array: Combining Spatial Annihilating and Manifold Separation

Abstract

In this paper, we address the problem of direction-of-arrival (DOA) estimation with the arbitrary array. The manifold separation technique (MST) is employed to transform the arbitrary array into a virtual array with Vandermonde manifold on which the spatial annihilating filter reconstruction method can be applied. When building the optimization problem for annihilating filter reconstruction, we propose the general solution modeling which can reduce the truncation error in MST to a negligible level. Finally, the spatial annihilating filter is reconstructed under the structural total least square (STLS) framework with the multiple measurement vectors structural total least norm (MMV-STLN) approach and the DOAs are estimated from the filter coefficients. Numerical simulations have verified the new proposed method adapts well to the low signal-to-noise ratio (SNR), limited snapshots and closely-spaced sources scenarios and can handle the coherent signals.