The Principles and Applications of Adaptive Filters:Adaptive Noise Cancelling, System Identification and Kalman Tracking

The digital signal processing field provides better solution of problems such as noise or interference cancellation, echo cancellation etc in various applications of communications, signal processing and biomedical. This is essential to remove noise or distortion from the signals. In digital signal processing, adaptive filtering is most significant region to remove noise or distortion. There are number of adaptive algorithms were developed for noise cancellation but LMS and RLS algorithms are more popular than others. This paper presents principles & application of adaptive filtering using different adaptive algorithms and simulation has done at MATLAB platform. This paper shows the concept of adaptive noise cancellation and implements the least mean square (LMS) and recursive least square (RLS) adaptive algorithms for noise cancellation. LMS and RLS algorithms are filter the noise from the input signal and gives noise free output signal. To identify the unknown plant, system modeling is also done in this paper. System identification is done by using LMS, NLMS & RLS Algorithms and also shows comparison graph between them. This paper also presents kalman tracking behavior using RLS. Simulation results shows that the performance of RLS has better adaptive noise cancellation as compared to that of LMS and also shows that RLS has minimum error than LMS & NLMS. The Graph of tracking behavior shows that actual & estimated signal are almost same.