A real-valued high-resolution coherent DOA estimation method with unknown source number

Abstract

When the signals are coherent and the number of sources is difficult to determine accurately, direction-of-arrival (DOA) estimation becomes challenging. In such scenario, the method proposed in this paper first reconstructs a Toeplitz matrix from the cross-correlation vectors of the array-received signals to perform decorrelation. This decorrelation technique preserves array aperture and facilitates DOA estimation. Subsequently, a new real-valued matrix is constructed using only the real and imaginary parts of the reconstructed matrix, instead of employing a unitary transformation. Based on this real-valued matrix, a synthetic spatial spectrum is formulated using subspace projection theory, requiring only a single matrix inversion and power operation, which improves computational efficiency. Simulation results and theoretical analysis demonstrate the effectiveness of the proposed method for estimating the DOAs of coherent sources in scenarios where the number of sources is unknown.