Abstract
Research summary
This paper deals with the simultaneous adaptive target detection and parameter estimation utilizing a Frequency Diverse Array Multiple-Input Multiple-Output (FDA-MIMO) radar. At the design stage, a linearized model of the received signal is derived, leveraging the centro-Hermitian (persymmetric) structure of the array manifold and considering potential steering vector mismatches related to the actual target parameters. Exploiting this model, the detection problem is formulated and tackled leveraging adaptive detection approaches, resorting to the Persymmetric Generalized Likelihood Ratio (PGLR) and Persymmetric Adaptive Matched Filter (PAMF), respectively. However, they demand the Maximum Likelihood (ML) estimates of target incremental range (i.e., the deviation of the target actual position from the center of the range bin) and angle. Two iterative methods are devised to solve the ML estimation problem. The former is based on Dinkelbach’s algorithm, which is capable of achieving the global optimum, whereas the latter is a fast-converging alternative approach relying on the Coordinate Descent (CD) framework. Numerical results highlight the effectiveness of the proposed simultaneous detection and estimation strategies, also in comparison with suitable benchmarks and counterparts.