A comparative study of two adaptive algorithms for suppression of a narrow-band interference

A comparative study of two adaptive algorithms which are available for suppression of a narrow-band interference is discussed. Two Notch Filters (NF) are presented in this paper: (1) Conventional Least Mean Square Filter (LMS-NF) and (2) Function Elimination Filter (FEF-NF). These algorithms present two different forms of linear prediction procedures, both of which are based on the LMS criterion. The first approach (LMS-NF) concerns the well-understood linear prediction. According to the second approach (FEF-NF), part of the filter coefficients are fixed and do not require any adjustment. The major part of the paper is devoted to the quantitative analysis of the considered algorithms. This analysis is carried out by studying the change in the Signal-to-Interference Ratio and the Bit-to-Bit Power Ratio due to a change in an additive noise level and in a power of a sinusoidal interference. The nature of the considered filters and their performances are demonstrated through digital computer simulation.