Projected gradient waveform design for fully adaptive radar STAP

We consider waveform design for radar space time adaptive processing (STAP), accounting for the waveform dependence of the clutter correlation matrix. It was shown previously in [1], that the joint problem of receiver weight vector optimization and radar waveform design is an intractable optimization problem, and constrained alternating minimization was proposed. In this paper, we propose projected gradient minimization. This minimization algorithm affords a numerically stable, lower computational complexity solution than the previously proposed alternating minimization. However, as a trade-off it results in larger iteration counts to achieve similar error variances of the STAP filter. Step size rules and the optimal step size for descent are derived analytically and validated via numerical simulations.