Improved Sparse Symmetric Arrays Design for Mixed Near-Field and Far-Field Source Localization

Symmetric uniform linear arrays (SULAs) are usually used for mixed far-field (FF) and near-field (NF) source localization. However, the low degrees of freedom (DOFs) of SULAs limits the accuracy of source localization and the number of sources that can be estimated. This issue can be solved by using sparse arrays, which are able to improve the DOFs and array aperture effectively. In this article, two sparse symmetric linear arrays (SSLAs) are designed for mixed source localization, which are named extended symmetric nested array (ESNA) and translation and transformation symmetric nested array (TTSNA). The ESNA configuration is made up of three sparse SULAs with certain interelement spacing, and it can obtain a larger number of unique lags than other SSLAs under the same number of array elements. The TTSNA is designed on the basis of nested arrays, making it possible to achieve a higher number of consecutive lags than other SSLAs, and the resultant virtual array elements is hole-free. Then, we give the closed-form expressions of the two proposed array configurations in terms of the number of unique and consecutive lags. Finally, simulation results prove that the proposed array configurations have better performance of mixed-field sources localization than other existing SSLAs.