Improved Conditional Posterior Cramér-Rao Lower Bound for Geometry-Dependent Bistatic Radar Measurement Uncertainty

Recently, the conditional posterior Cramér-Rao lower bound (CPCRLB) is proposed by incorporating sensor acquired target's historical measurements into its recursion derivation, and provides an online bound to evaluate the lowest mean square error (MSE) an estimator can achieve in theory more faithfully. Whereas, the existing CPCRLB completely neglects the fact that radar measurement uncertainty is significantly dependent on the target-to-radar (T2R) geometry, and discards available target information that unavoidably results in bound deterioration. Motivated by the fact that radar measurement uncertainty is significantly dependent on the T2R geometry, this paper formulates the geometry-dependent target measurement uncertainty model for bistatic radar system and derives an improved CPCRLB (ICPCRLB) with imperfect detection based on this model. Because of fully accounting for the impact of geometry-dependent target measurement uncertainty on the conditional bound, the proposed ICPCRLB is a more accurate MSE bound than the existing CPCRLB. Simulation results demonstrate the superiority of the proposed ICPCRLB over existing state-of-the-art bounds.