Nonlinear Interference Cancelling In Biomedical Systems

While the linear FIR adaptive filtering is very useful and simple for interference cancelling, there are many practical situations where the use of a nonlinear model is more realistic. In this paper a method for nonlinear system identification is briefly described. A new interference canceller scheme based on this method is also discussed. This scheme is applied to two different biomedical signals: Respiratory artifacts cancelling in pulmonary pressure transients and ECG interference cancelling in diaphragmatic muscle signals. Good results are obtained making this scheme attractive also for other applications.