Triple extended coprime array for single‐snapshot DoA estimation

Abstract

AbstractThe direction‐of‐arrival (DoA) estimation methods based on coprime array (CA) largely rely on having a sufficient number of snapshots. However, under single‐snapshot conditions typical of vehicle‐mounted imaging radar, CA‐based DoA estimation often suffers from ambiguity leakage, resulting in significant degradation in performance. This paper thoroughly analyses the mechanism of ambiguity leakage in CA by re‐deriving the covariance matrix model under single‐snapshot conditions. To suppress the ambiguity leakage, a triple extended coprime array composed of three pairwise coprime sub‐arrays enhanced coprime constraint is designed. Building on this design, a DoA estimation algorithm based on an iterative adaptive approach is proposed. By leveraging the power spectrum distribution from iterative adaptive approach, a reduced‐noise signal is constructed by retaining only the high‐energy components. Next, the noise subspace using a multi‐stage Wiener filter, and finally, achieve super‐resolution DoA estimation through multiple signal classification spectrum analysis.