2-D Off-grid DOA Estimation for Parallel Coprime Array on Moving Platform

In this paper, we develop a new array model based on the parallel coprime array on moving platform. With the array motion, a synthetic array composed of the original parallel coprime array and its shifted version can be obtained, which has more unique and consecutive lags in the difference coarray and thus owns higher degrees of freedom than the original parallel coprime array. Furthermore, to solve the grid mismatch problem caused by the discretization of angular space, we also propose a 2-D off-grid DOA estimation method. First, the cross-covariance matrix of the synthetic array is calculated and vectorized to get the virtual received signal model. Then, grid bias is introduced to this model and an off-grid model is thus built. Finally, by solving the optimization problem constructed based on the off-grid model, we can obtain an accurate estimate for the 1-D angle. The other dimensional angle can be acquired by the least square method. In the end, simulation results confirm the superiority and effectiveness of the proposed method.