This paper presents an optimized FIR filter implementation by using the method of improving coefficient precision at reasonable cost. The comparable results of traditional fix-point implementation and the optimization method show that the high-precision FIR filter design method is universal and easy to implement. Plus the employment of multi-stage pipelining and parallel structure, FIR filter performs higher operating frequency. Let’s take a 32-order lowpass FIR filter as an example, original coefficients are generated on MATLAB, and translated into optimized coefficients according to the optimization method. The functional simulations verify the effective performance, while the synthesis is carried out to analyze the utilization of resources and maximum frequency.
An Optimal FIR Filter Design Method Using H/W Complexity Estimation