Adaptive Step‐Size PtNLMS Algorithms
Kevin T. Wagner, Milos I. Doroslovacki · 2013
This chapter presents an adaptation of the μ-law for compression of weight estimates using the output square error. Additionally, it explains a simplification of the adaptive μ-law algorithm, which approximates the logarithmic function with a piecewise linear function. A section presents the MSE versus iteration of the ASPNLMS, AMPNLMS and AEPNLMS algorithms. The chapter discusses the three new PtNLMS algorithms. The AEPNLMS algorithm is a modification of the EPNLMS algorithm with an adaptive μ parameter. This algorithm shows improvement with respect to the EPNLMS algorithm. The final algorithm, ASPNLMS, is a simplification of the AMPNLMS and AEPNLMS algorithms. The ASPNLMS algorithm avoids the usage of the logarithm function by approximating the logarithm with a two-segment linear function. This algorithm reduces the computational complexity and can offer superior convergence performance to the AMPNLMS and AEPNLMS algorithms. Controlled Vocabulary Terms computability; error analysis; least mean squares methods