Dynamic Coefficient Quantization Digital FIR Filter

Abstract

The rationale for using the quantization of the coefficients of a digital FIR filter in the concept of dynamic programming is given as a process of stepwise quantization of the coefficients with their discrete optimization at each step according to the objective function common to the entire quantization process. An algorithm for step-by-step dynamic quantization is considered by methods of integer nonlinear programming, taking into account the given signal scaling. The effectiveness of this approach is illustrated by the example of dynamic quantization of the coefficients of a cascaded high-pass FIR filter with a minimum bit depth of representation of integer coefficients.