ABSTRACT The existing tensor‐based methods only exploit one or step‐by‐step tensor decomposition to estimate the array manifolds of the monostatic frequency diverse array multiple‐input multiple‐output (FDA‐MIMO) radar, and cannot simultaneously exploit different tensor decompositions to effectively represent data features. Thus, this letter proposes a hybrid tensor ring and CANDECOMP/PARAFAC (CP) decomposition for FDA‐MIMO radar, which uses tensor ring decomposition to capture the multi‐way correlations and CP decomposition to extract identifiable array manifold factors, leading to more accurate parameter estimation.
Yue et al. (Thu,) studied this question.