An Improved Model-Based Wald Detector for Range-Spread Targets in Heterogeneous Clutter Environments

Autoregressive (AR) modeling of heterogeneous clutter can improve the probability of detecting range-spread targets, but the detector performance diminishes with fewer pulses due to the loss of samples equivalent to the AR order. To mitigate this performance decline, this paper presents an improved model-based Wald detector for range-spread targets in heterogeneous clutter environments, under the assumption of a known clutter covariance matrix based on the Wald test. Furthermore, the covariance matrix is reconstructed using the lower triangular matrix and real symmetric matrix, both of which are composed of the estimated AR parameters. In addition, considering the varying sea clutter under different sea states and across different range cells, an adaptive detector system consisting of detectors in heterogeneous and homogeneous environments is designed. Based on simulated and real clutter data, we demonstrate that the improved detector outperforms the AR model-based detectors with fewer pulses, and that the adaptive detector system can effectively select the optimal detector for the current clutter environment, ensuring high detection performance.