Robust Incremental Nonlinear Channel Estimation based on Spline Adaptive Filtering

Linear modelling of the communication channel is an oversimplified assumption, as there may be nonlinear distortions in the transmitted symbol. Spline adaptive filtering based on least mean square criteria (SAF-LMS) is used by researchers to estimate the nonlinear system in additive white Gaussian noise (AWGN). However, the failure of SAF-LMS in an impulsive noise environment prompts us to research further on robust cost functions. Hence a robust spline adaptive algorithm is proposed to estimate the nonlinear channel. Robustness comes with a toll on the convergence rate. To make the algorithm energy-efficient, distributed technique with an incremental way of co-operation, where multiple nodes are connected in a ring topology, is incorporated to estimate the channel. The simulation results show the robustness of the algorithm against impulsive noise.