Performance analysis of adaptive lattice filters for impulsive signals

The existence of impulsive signals with /spl alpha/-stable non-Gaussian distributions has already been reported in some applications. A new adaptive lattice algorithm is proposed for adaptive filtering of /spl alpha/-stable AR processes. The performance of this algorithm is compared to that of other stochastic gradient-based algorithms for stable AR processes for different /spl alpha/'s. It is shown that the proposed algorithm achieves a superior convergence speed with respect to the other algorithms.