Unambiguous Detection of Migrating Targets with Wideband Radar in Gaussian Clutter

In this paper the problem of unambiguous moving target detection using wideband radar is considered. Doppler ambiguities, which are present in the low pulse repetition frequency mode, are transformed into ambiguous sidelobes of targets and clutter using range migration effect. However, the level of these sidelobes is typically high, such that standard detectors suffer from false alarms generated by the ambiguous sidelobes of real targets. To remove these false alarms two new detectors assuming Gaussian distribution of clutter and exploiting high resolution spectral estimation are developed. These detectors are based on spectral estimation obtained with non-parametric Iterative Adaptive Approach from spectrum analysis and satisfy bounded false alarm condition. Approach to reduce the number of reference cells for covariance matrix estimation is discussed. The benefits of the proposed detectors are demonstrated via numerical simulations showing improvement over existing algorithms.