On the use of HRR data to improve target kinematics estimation: CRLB computation and comparison with simulated results

An innovative approach to jointly estimate and compensate the radial velocity and acceleration of non-cooperative targets in HRR (High Range Resolution) processing for a frequency stepped radar has been proposed in [1] and [6]. The estimation is performed using different approaches mainly based on the maximization or the minimization of a proper functional (contrast, correlation-like or entropy) of the reconstructed profile. The partial improvement found for the target velocity and acceleration estimation indicates that the three analyzed “cost functions” don't reach ideal performance. Scope of this paper is to quantify the best theoretical achievable performance by means of the Cramer-Rao Lower Bounds (CRLB) and compare the obtained results in [1] and [2] with the CRLB.