Fast Direction-of-Arrival Estimation in Coprime Arrays

We present a fast direction-of-arrival (DOA) estimation algorithm for coprime array structures. In many practical applications of radar signals processing, reducing the redundancy inherent in linear arrays is advantageous. Arrays employing a coprime-pair configuration offer an increased number of degrees of freedom and permit the detection of more sources than sensors. Traditionally, this is achieved using conventional subspace-based methods. While these methods, such as MUltiple Signal Classification (MUSIC), are capable of high-resolution, the computational complexity of these algorithms is a significant hindrance to their practical implementation as the number of sensors increases. Furthermore, they must resort to spatial smoothing in order to calculate the covariance matrix. We, on the other hand, propose a fast iterative interpolated beamforming algorithm that has a computational complexity of the same order as the fast Fourier transform (FFT). The proposed approach is capable of delivering high fidelity DOA estimates that outperform the traditional high resolution methods. Simulation results demonstrate the DOA estimation performance of the technique.