Nonlinear Adaptive System Identification Based on Wiener Models (Part 4)
Having developed the nonlinear Wiener LMS-type adaptive filtering algorithms in the previous three chapters, we show in this chapter that the RLS-type algorithm can be applied for the nonlinear Wiener model too. The trade-off is between convergence rate and computational complexity. In addition, for practical VLSI implementation the inverse QR decomposition for the recursive least squares (RLS-type) algorithm can be combined with the nonlinear Wiener model to achieve a highly efficient systolic array architecture.
KeywordsAdaptive Filter Subset Selection Systolic Array Recursive Little Square Bilinear System
Unable to display preview. Download preview PDF.