Convergence analysis of stochastically-constrained sample matrix inversion algorithms

It has been recently demonstrated by both computer simulations and real data processing that multi-interference signal environments with different types of interference stationarity can be adequately treated by the newly proposed stochastically-constrained adaptive algorithm. This signal processing approach is evidently the prototype of a new class of adaptive algorithms, whose convergence properties are analytically and numerically examined in this paper. Interference scenario which reflect the main features of typical HF radar applications are presented; these demonstrate both the high efficiency of the approach described and the accuracy of the derived analysis.