Joint DOD and DOA Estimation via Coupled CPD with Sparse MIMO Radar for IIoT

Coarray tensor-based joint direction-of-departure (DOD) and direction-of-arrival (DOA) estimation with sparse multiple-input multiple-output (MIMO) radar for industrial internet of things (IIoT) has been a heated topic, where the canonical polyadic decomposition (CPD) is utilized to retrieve angle information. However, the conventional CPD approach ignores the spatial relevance among the coarray tensors corresponding to the collocated transmitters and receivers of the sparse MIMO radar. In this paper, we propose a coupled CPD-based joint DOD and DOA estimation method for the sparse MIMO radar in IIoT. Firstly, multiple tensors are modeled to preserve the multi-dimensional signal structure of the sparse MIMO radar. Then, tensor-based spatially smoothed coarray statistics corresponding to augmented virtual transmit and receive arrays are derived. After that, a segmented coarray tensorization approach is proposed to formulate multiple coupled coarray tensors. To make full use of the relevance among these coarray tensors, they are jointly decomposed via coupled CPD for joint DOD and DOA estimation with a high accuracy and resolution. The effectiveness of the proposed method is verified by simulations.