Performance Research And Application Of Adaptive Filtering Algorithm

In the current information age,adaptive signal processing is an important branch of information processing,is widely used in communication,radar,sonar,industrial control,earthquake prediction and biomedical fields.Among them,the adaptive filter is an important part of the adaptive signal processing technology,and the adaptive algorithm is the most important factor in the adaptive filter design,therefore,the research of adaptive filtering algorithm is one of the most active topics in signal and information processing.Because the performance of the algorithm will directly affect the performance of the filter,so seek the faster convergence speed,lower steady-state error,lower computational complexity,better tracking ability of adaptive filtering algorithm has been the goal of many scholars.This paper describes the working principle of the adaptive filter and the common filtering theory and algorithm.Mainly introduces the LMS algorithm based on Wiener filtering,and in view of the shortcomings of traditional LMS algorithm,combined with the literature,makes a further study on the two aspects of variable step size LMS algorithm and convex combination filtering algorithm.Something specific in the following two parts.In the aspects of the variable step size LMS algorithm,study the common variable step size algorithm,summarized the basic principle of variable step size LMS algorithm design.According to working principle,combined with hyperbolic secant function,introduce adjustment parameters,established a new type of nonlinear function between the step factor and the error signal,proposed a variable step size LMS algorithm based on hyperbolic secant function.In the same simulation environment,this new algorithm is compared with other known algorithms,the results show that the new algorithm has a significant increase in the convergence rate under the premise that the steady-state error does not change.Finally,the new variable step size LMS algorithm is applied in the application of adaptive noise canceller,and get a satisfactory effect.In the aspects of the convex combination filtering algorithm,improved for the shortcomings of CLMS algorithm in poor flexibility and traceability: in the traditional convex combination filter,introduced the variable step size LMS algorithm replace the original large step LMS algorithm,propose a new variable step size convex combination algorithm,prove the feasibility of the algorithm by theory andexperiment.In order to reflect the superiority of the algorithm,the new algorithm is simulated with the traditional convex combination algorithm and the other variable step size convex combination algorithm in the same identification system with system mutation.Based on the experimental results,the new algorithm has the best filtering effect,and the tracking performance is greatly improved.