2-D DOA estimation of coherent signals with uniform circular array based on compressed sensing and genetic algorithm

Unlike the uniform linear array (ULA), the beam of uniform circular array (UCA) can scan azimuthally through 360° and the resolution is the same for any angle. What is more, the UCA can realize two-dimension (2-D) direction of arrival (DOA) estimation of coherent signals. However, most mature algorithms were designed for ULA. Since the array manifold matrix of UCA does not satisfy the Van der Monde structure, the decoherence algorithms for ULA cannot be applied to UCA directly. Therefore, in this paper, we propose a novel method to estimate the 2-D DOA of coherent signals with UCA based on compressed sensing (CS) and genetic algorithm (GA). The CS algorithm is used first to demodulate the coherent signals, the angle information of the target is preliminarily obtained. Then, the spectrum peak search is performed near each estimated angle to improve the estimation accuracy, meanwhile, the GA is employed to accelerate the convergence speed and reduce the calculation overhead of search. Finally, it outputs an accurate 2-D DOA value. The method breaks through the limitations of the array structure, and successfully realizes the estimation of the pitch angle and the horizontal angle simultaneously. Numerical simulations are conducted to verify that the proposed method shows good performance in 2-D DOA estimation of coherent signals.