Adaptive Consistent Parameter Estimation for Non-Gaussian MA Processes Using 4th Order Cumulant

摘要

This paper addresses the problem of adaptively estimating the consistent parameters for non-Gaussian nonminimum MA processes with symmetric PDF using the fourth order cumulant of the underlying processes. The processes may be corrupted by additive noise. The proposed adaptive algorithm is derived by using the unique-solution e-quation, and therefore can provide the consistent estimates of processes' parameters. Some experiment results are provided to show the performance of our new adaptive algorithm.