Abstract: Adaptive filtering is an important signal processing area that has wide applications in communications, control, and biomedical engineering fields. The adaptive noise cancelling, adaptive equalization of data transmission channels and adaptive antenna arrays are some of the examples of such applications. Adaptive filtering consists of a digital filter whose weights are controlled by an adaptive algorithm so on minimize the difference between the filter output and a reference signal consistent with some criterion. The character of the reference signal depends on the considered application. There are two main measures for evaluating the performance of an adaptive filter: the convergence rate and therefore the steady state mean square error. In practical applications, it's desired to maximize the convergence rate and minimize the steady state mean square error. There is a conflict between these requirements. Several adaptive algorithms have been developed so as to yield a good compromise between these requirements. The important adaptive algorithms are Sample Matrix Inversion (SMI), Least Squares (LS), and Recursive Least Squares (RLS) algorithm. The main objective of this project is the implementation of LMS and RLS (Recursive Least Square) adaptive filters algorithm using Xilinx system generator. The simulation of the models will be done in Matlab and Simulink for the efficient verification of the algorithm. The core RLS and LMS adaptive filtes and its basic components block will be developed in Xilinx System Generator and implementation in Xilinx FPGA.
Keywords: Least mean square algorithm LMS, Recursive Least Square algorithm RLS, Xilinx system generator XSG, simulink, Spartan -3