Step value-variable LMS (Least Mean Square) self-adaptation filtering algorithm and filter

Abstract

The invention relates to the technical field of digital signal treatment, in particular to a step value-variable LMS (Least Mean Square) self-adaptation filtering algorithm and a step value-variable LMS (Least Mean Square) self-adaptation filter. According to the invention, since a step value capable of changing according to filtering stages can be supplied, a higher convergence rate and a smaller system stability error can be synchronously obtained, which shows that in the initial stage of self-adaptation filtering, a larger step value is provided, thereby obtaining a higher convergence rate, and a smaller step value is provided when the self-adaptation filtering approaches a stable state, so that a smaller stable state error value can be obtained.

Description

technical field [0001] The invention relates to the technical field of digital signal processing, in particular to a variable-step LMS (Least Mean Square, least mean square) adaptive filtering algorithm and filter. Background technique [0002] Adaptive filter has always been one of the research hotspots in the field of signal processing. After years of development, it has been widely used in digital communication, radar, sonar, seismology, navigation system, biomedicine and industrial control and other fields. [0003] The most widely used adaptive algorithm is the Least Mean Square (LMS, Least Mean Square) algorithm. The LMS algorithm is a search algorithm that simplifies the calculation of the gradient vector by properly adjusting the objective function. Due to its computational simplicity, the LMS algorithm and others related to it have been widely used in various applications of adaptive filtering. The basic idea of ​​the LMS algorithm is to adjust the weight coefficie...

AI Technical Summary

Problems solved by technology

[0009] In the LMS filtering algorithm in the prior art, the algorithm step size is a fixed value, and a better filter convergence time and stable error value cannot be obtained at the same time

Specifically, when the step size u is set larger, the filter can quickly reach the convergence state, but if the filter is close to a stable state, the stable error value will be larger, which affects the error performance of the system; when the step size u is set smaller , the filter can obtain a small stable error value. If the convergence time is long in the initial stage of adaptive filtering, it is necessary to increase the training sequence length of the algorithm

Images