MADALINE neural network with truncated momentum for LTV MIMO system identification

Presented in this paper is a new version of the Multi-ADAptive LINear Element (MADALINE) neural network for Online System identification of linear time-varying (LTV) Multi-Input Multi-Output (MIMO) systems. A truncated momentum term is used in the learning algorithm for the purpose of reducing fluctuation while sudden parameter change happens thus offers a smoother transition in tracking the parameter. Based on the input output polynomial model, which can be easily transformed into the row canonical state space model, Tapped delay lines are introduced, so the MADALINE becomes recurrent in nature and thus is suitable for parameter estimation of such systems. The MADALINE can then be setup under the assumption that the system structure is known in advance. The estimated parameters are obtained as the weights of trained individual neurons of the MADALINE. The method is implemented in MATLAB and simulation study was then performed on a few well known examples. Simulation results show that the algorithms offer satisfactory performance. This work is based on our previous work on Multi-Input Multi-Output systems´ identification [18].