On bounds for the coefficient search space with application to discrete optimization in a class of wave digital filters

A method of generating bounds for the multiplier coefficient search space for discrete optimization in a class of wave digital filters is proposed. The method is based on a region of deviation allowed for the poles and zeros of the characteristic function of the filter. The method has a computational advantage in comparison with methods based on large change sensitivity tools. Although rough bounds are generated, they reduce drastically the coefficient search space.<
>