Optimized Sparse Nested Arrays for DoA Estimation of Non-circular Signals

The recent criteria for designing sparse arrays used in many applications are the number of degrees of freedom (DoFs) and mutual coupling. In this paper, two optimized sparse nested arrays, called OSNA-I and OSNA-II, are proposed for the direction of arrival (DoA) estimation of non-circular signals. The proposed structures aim to enhance the DoFs and mitigate mutual coupling. The OSNA-I and OSNA-II are obtained by rearranging physical sensors of the conventional nested array. Specifically, for the proposed OSNA-I, its dense uniform linear array (ULA) is exactly the same as the dense ULA of the conventional nested array, while the physical sensors of its sparse ULA are located at . For the proposed OSNA-II, it is composed of two subarrays plus one separate sensor that is appropriately placed. The OSNA-II keeps the dense array of the conventional nested array untouched except the sensor at being removed, while its sparse array is deployed according to The separate sensor is set apart with the rightmost sensor of the sparse array by a space of . The closed-form expressions are derived for the actual sensor locations, number of uniform DoFs and available DoFs, respectively. Besides, it is also proven that OSNA-I and OSNA-II have small weight values due to their sparse structures, which means that OSNA-I and OSNA-II can achieve a good balance between DoFs and weight values. Simulation results demonstrate the superiorities of the proposed structures in terms of robustness and estimation accuracy.