Output details

A New Robust Variable Step-Size NLMS Algorithm

We introduced a new framework for designing robust adaptive filters based on the optimisation of a cost function with a time-dependent constraint. The extra advantage is the ability to reach a low steady-state error without compromising the speed of convergence. Moreover, a theoretical model was introduced, having the potential to be used in many different adaptive filters to have accurate predictions of their transient and steady-state behaviour. Based on this framework another four papers were published introducing new algorithms suitable for different applications. This work set the basis for a new book I have recently co-authored (Springer, 2012).