Eigen-space analysis of constrained adaptive radar waveform design

The trade-space of performance gain versus implementation cost is an important topic in transmit-adaptive radar processing. For signal design, one must be concerned with the signal-to-noise and interference ratio (SINR) gain, hardware needs for waveform synthesis (including bandwidth requirements, power constraints, etc.), and the shape of the ambiguity function when compared against a traditional non-adaptive signal (e.g. linear frequency modulated signal). We investigate the attendant performance degradation when constraints are imposed relating to the aforementioned issues. We explore the effect of these constraints on the objective function performance as a function of signal length, interference complexity and training data support. Numerical results using Monte Carlo simulations are presented.