Zero tracking algorithms: from stationary to nonstationary environments

In many situations, the plant representation in terms of poles and zeros may provide a more appropriate description of a process than other widely used representations. We propose an autocorrelation-based zero tracking algorithm (ZTA) to deals with stationary process. For the time varying process, a weighted functional series modelling (FSM) of a time varying system´s poles and zeros is proposed. This approach is based on the Gauss-Newton recursive prediction error method with an exact expression for the gradient