Performance analysis of adaptive algorithms and enhancement using Kalman filter

A new platform for designing robust adaptive filter is introduced. An adaptive filter is a filter that adjusts its transfer function according to an optimizing adaptive algorithm. The efficiency of the adaptive algorithm being used plays a key role in the working of the adaptive filter. The Least Mean square (LMS) and the Normalized Least Mean square (NLMS) adaptive algorithms are studied. The core part of this research is to use the theory of Kalman filter and use it in adaptive filtering process. The adaptive filtering problem can be updated to a new theory of state estimation problem. The main objective of the research is to evaluate and characterize the efficiency of the adaptive algorithms being used in the adaptive filtering process. The adaptive filtering process will be carried out using different adaptive algorithms and its efficiency is measured in terms of filter convergence speed and the variation in the power of the error signal with changes in the input signal power obtained during the adaptation process. A Kalman based Normalized Least mean square algorithm which is developed outperforms the existing Least Mean square (LMS) and Normalized Least Mean square (NLMS) Algorithms. The simulations are carried out by using MATLAB.