.png){kind=icon}
.png){kind=icon}
Return
Lightweight Coprime Array Virtualization Method Based on Polynomial Interpolation
.png){kind=icon}
PRE
AI
DOI：10.1109/taes.2026.3702656.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Multitarget sensing and detection is an important scientific proposition for radar signal processing, and direction-of-arrival (DOA) estimation is a key subsequent step after target detection. In order to reduce hardware complexity, aperture virtualization is usually used to improve the performance of the sparse multiple-input multiple-output (MIMO) radar which has less physical channels. However, the existence of holes in the virtualization of the sparse array will lead to the loss of some array element information, degrading radar's detection performance. Existing interpolation strategies commonly recover these missing lags through convex optimization, improving estimation capability at the expense of substantial computational cost. In this article, we propose an array virtualization method based on polynomial interpolation to optimize radar's multitarget DOA estimation performance with reduced computational complexity. In this method, the received signals will be preprocessed under the idea of power equalization. Subsequently, the proposed polynomial interpolation algorithm is introduced to fill the holes in coprime array, a typical kind of sparse radar array. The outputs are second-order statistics, which are related to the virtual measurements under the Hermitian positive semidefinite Toeplitz condition. Consequently, the virtual interpolation signals covariance matrix is yielded by the Toeplitz covariance matrix reconstruction approach, making it suitable for all the mainstream radar DOA estimation method. Meanwhile, a lightweight coprime array virtualization method is proposed, which can further enhance DOF performance with the proposed array configuration. Both simulation and hardware validation results demonstrate the improvement on both of multitarget DOA estimation performance and computational complexity, making it a practical solution for diverse lightweight radar platforms.
Keywords:
Array virtualization
direction-of-arrival (DOA) estimation
polynomial interpolation
sensing radar
Journal
IF:
5.7
Papers:
676
Citations:
2.4W
View more organizations