A super transformed nested array with reduced mutual coupling for direction of arrival estimation of non‐circular signals

Fengtong Mei, Haiyun Xu, WeiJia Cui, Chunxiao Jian, Jian Zhang
2022 IET radar, sonar & navigation  
Recently, sparse linear arrays have aroused significant attention because they can provide increased degrees of freedom (DOFs) and enlarged inter-element spacing, which are the critical factors to improve the accuracy of direction of arrival (DOA) estimation. In this paper, to alleviate the high mutual coupling caused by the dense part of transformed nested arrays (TNAs), we present a super transformed nested array (STNA) for the DOA estimation of non-circular signals. It can be obtained by
more » ... ving some sensors from the dense part of transformed nested arrays and relocating them appropriately. The proposed array structure possesses a higher DOF and a sparser physical structure, resulting in less mutual coupling compared to TNA. For the proposed array structure, we provide theoretical proof of the DOFs and deduce the weight functions. Numerical simulations are conducted to demonstrate the advantages of the STNA over existing sparse arrays. K E Y W O R D S degree of freedom, direction of arrival (DOA), mutual coupling, nested arrays, non-circular signal | INTRODUCTION Direction-of-arrival (DOA), as an important aspect of array signal processing, is widely used in scenarios such as radar, sonar, and wireless communication [1] [2] [3] . A uniform linear array (ULA) with M sensors based on the subspace algorithm such as multiple signal classification (MUSIC) [4] can only detect M − 1 sources, which means that the more the sources, the more the sensors required. Moreover, the inter-element spacing of ULAs are not large than half wavelength of incident sources, which will cause severe mutual coupling effects. Recently, sparse arrays have attracted great interest because they can provide larger inter-element spacing and higher number of degrees of freedom (DOFs) than ULAs with the same number of sensors. The minimum redundant array (MRA) is proposed in Ref. [5] , which can generate the maximum number of consecutive lags based on the difference co-array (DCA). But its design requires very complicated computer searches because of the lack of analytical expressions of sensor positions. The coprime arrays (CPAs) [6] have exact expressions of sensor positions, making array construction easier than that for MRA. In Ref. [7] , the generalized coprime array (GCA) is proposed, which includes two formation, that is, CPAs with compressed inter-element spacing (CACIS) and CPAs with displaced sub-arrays (CADiS). Compared with CPA, CACIS can provide an increased DOFs, while CADiS possesses a larger inter-element spacing. However, there are some holes in the DCA of the GCA, leading to the decrease in the number of DOFs. Different from GCA, the nested arrays such as the nested array (NA) [8] , the super nested array (SPNA) [9] , and the augmented nested array (ANA) [10] can provide hole-free DCAs. The NA consists of a dense ULA and a sparse ULA, where the dense part is sensitive to mutual coupling. The SPNA can be obtained by extracting some sensors in the dense sub-array of NA and then shifting them, which keeps the same advantage of NA and reduces the mutual coupling effect. The ANA, generated by dividing the dense sub-array of NA and placing them separately, achieves a higher number of DOFs than SPNA. Based on these sparse arrays, many novel DOA estimation methods, such as spatial smoothing subspace MUSIC [11] and sparse recovery algorithm [12] [13] [14] , have been proposed. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
doi:10.1049/rsn2.12221 fatcat:mdggioi6urfphdvlmkfvrbkgua