Range-Angle Decoupling and Estimation for FDA-MIMO Radar via Atomic Norm Minimization and Accelerated Proximal Gradient

Frequency diverse array multiple-input multiple-output (FDA-MIMO) radar can offer the capability of range-angle-dependent beampattern and enjoy the advantage of effectively resolving the targets closely spaced in the same angle cell but different range cells. However, traditional subspace methods for range-angle estimation fail to efficiently work in a limited number of snapshots and coherent targets. In addition, the coupling between range and angle may lead to the degradation of estimation accuracy with ambiguity problem. In this letter, a gridless compressed sensing-based algorithm is proposed to joint estimate range-angle for FDA-MIMO radar. First, a decoupling model is presented to separate the range and angle from each other. Then, a 2D atomic norm minimization (ANM) problem for range-angle estimation is formulated and transformed into a semi-definite programming (SDP) problem with convex relaxation. Finally, a computationally efficient estimation algorithm via accelerated proximal gradient (APG) is developed to solve the SDP problem. Numerical simulations are conducted to illustrate the superior performance of the proposed algorithm.