A Technique for Compensating the Filter Performance by a Fictitious Noise

This technical brief is concerned with an adaptive technique for compensating the filter performance in an erroneous system model. The degradation of the performance is covered by adding a fictitious input noise and adaptively estimating the statistics. The estimates of the statistics are evaluated by exponentially ageing the old information about them. Numerical results show that the proposed approach is much more improved than a suboptimal filter.