Online identification of a system with shared band noises

This paper proposes an online framework to dynamically model the impulse-response function (IRF) of a system having shared-band noises, to facilitate the output prediction in real-time. The online independent-component analysis is performed to un-mix the measured signal. The automatic recognition of the anticipated IRF, amongst the unmixed signals, is achieved by proposing a peak-detection-&-correlation technique. A mathematical model of the acquired IRF is, then, dynamically identified by the subspace-based state-space method. The validity of the proposed methodology is demonstrated by a simulation study where the anticipated IRF is blindly identified and then modelled with an accuracy of 92%. The framework has the potential of online modelling of a system having shared-band noises.