Thesis Archive
A Hardware Model for an Adaptive Finite Impulse Response Filter (2007)
Patrick Chris H. Oliva, Dino Martin C. Manalo V and Ryan Carlos B. Dy, advised by Engr. Roderick Yap
Abstract:
-The system used a Transversal filter where the Least Mean Square Algorithm operates; and the system is implemented in an XC4000E HQ240 Field Programmable Gate Array. The application of the system is reducing the noise in a speech signal. Before determining the number of tap weights and the value of the step size parameter ? of the system, the group conducted the study about the operation of the Least Mean Square Algorithm. The two input signals are obtained and buffered from a single computer through its parallel port. The system is tested using an oscilloscope to visually see the Estimation Signal or the Estimation Error. The digital output was fed to a PC speaker to hear the Estimation Signal or Estimation Error as well. The simulation results of Xilinx Foundation Series 2.1 are compared to the results of MATLAB 6.5, used a basis for comparing and determining the optimum number of tap weights and step size parameter ? of the system.