Chen Y.-PCHUN-LIN LIU2023-06-092023-06-09202215206149https://www.scopus.com/inward/record.uri?eid=2-s2.0-85131256219&doi=10.1109%2fICASSP43922.2022.9746771&partnerID=40&md5=b456c603f3cda72fe626020275aca5a1https://scholars.lib.ntu.edu.tw/handle/123456789/632112Sparse arrays with fourth-order cumulant processing can identify up to O(N4) source directions with only N physical sensors. To achieve this property, the fourth-order difference co-array is preferable to own a contiguous segment with no holes. However, existing sparse array designs either have holes in the co-array of size O(N4) or own fewer than O(N4) elements in the hole-free co-array. This paper proposes the half inverted nested array (HINA), which consists of a nested array and an inverted, scaled, and shifted nested array. By maximizing the size of the co-array, it can be shown that HINA with the optimal parameters possesses a hole-free fourth-order difference co-array of size O(N4). Numerical examples demonstrate the improved DOA estimation performance of HINA. © 2022 IEEEdirection of arrival estimation; fourth-order difference co-arrays; nested array; Sparse arraysconference paper10.1109/ICASSP43922.2022.97467712-s2.0-85131256219