Low-complexity Direction-of-arrival Estimation of Coprime Array based on Noise Subspace Reconstruction

Coprime array is a new promising sparse array which has been widely used in direction-of-arrival (DOA) estimation. The sparse coprime array brings many advantages such as large array aperture, high degree of freedom and high precision DOA estimations, while the complexity is high because of the discontinuous and sparse physical array sensors. In order to reduce the computational complexity of DOA estimation based on coprime array in overdetermination condition, a low complexity DOA estimation based on noise space reconstruction is presented in this paper. This method firstly calculates the covariance matrix of the received sources, and then the noise subspace can be obtained by eigenvalue decomposition. The sparse non-uniform array has been transformed into continuous virtual array with the help of noise subspace reconstruction. Finally, the low complexity DOA estimation can be realized by root MUSIC method. The compared analysis of complexity is given. The simulation results show that under the same conditions, the DOA method proposed can achieve the same estimation accuracy of spectral search MUSIC method with a low complexity.