A new approach to near optimum MTI filter design using adaptive techniques

Presents a new approach to a near optimum MTI filter design using adaptive techniques. The resulting scheme consists of an adaptive whitening filter for radar clutter followed by a likelihood ratio test (LRT) for target signal in white noise. A complete procedure is developed for the whitening filter design. The LRT may operate either with or without a clutter power map. The detection performance of the new scheme is evaluated using real radar data. When operating with a clutter map, it provides 7.5 dB average performance improvement over the innovations-based detection algorithm (IBDA) for ground clutter-dominated data and 5 dB improvement for weather clutter. And the latter has been demonstrated to be superior to the classical moving-target-detection (MTD) by 3 to 5 dB.<
>