Efficient algorithm to design quadrature mirror filters using multi-band weighted Lp error criteria

Abstract

An efficient iterative algorithm is presented in this paper to design quadrature mirror filters by minimizing the p-th power of the error function, where p may varied in different frequency band. Our algorithm incorporates the iterative re-weighted least squares technique. The filter bank design is approximated as a weighted least-squares quadratic programming problem. Good designs are obtained typically in a few iterations. The algorithm designs quadrature mirror filters which satisfy mixed Lp error criteria in the overall filter bank frequency responses and the filters stop-band responses.