ON ROBUST AML IDENTIFICATION ALGORITHMS

Strong consistency results for a class of nonlinear approximate maximu m likelihood algorithms for robust system identification are developed , where the system is assumed to be of the ARMAX form. The analysis us es the Martingale results, and strong consistency is shown to hold und er a new assumption, representing a generalization of the strictly pos itive-real condition. Arguments are also given for using Huber's nonli nearity, in order to reduce the influence of outliers in practice.