On the Optimization of MTI Clutter Rejection

In this paper is formulated the problem of optimization of the improvement factor of a nonrecursive MTI by minimization of a quadratic form. The minimum normalized clutter output (a reciprocal of the average improvement factor) is the minimum eigenvalue of this quadratic form, and the corresponding eigenvector is the optimal weight to be used in this filter. By use of classic matrix theory, some properties of this MTI improvement factor are shown, namely, that it is bounded and is a monotonic function of the clutter spectrum variance. Also discussed is the limit of an MTI system having a large number of cancellers. Finally, the problem of a staggered-PRF MTI filter is examined, for which it is shown that its improvement factor is bounded by two equivalent constant-PRF MTI systems. One of these systems has a PRF equal to the lowest PRF of the staggered-PRF system, while the other has a PRF equal to the highest PRF of the staggered system.