Synthesis of low coefficient sensitivity digital filters using genetic programming

This paper proposes a new approach to the synthesis of low coefficient sensitivity digital filters using Genetic Programming (GP). GP is applied to the synthesis problem by establishing a mapping between the Sexpressions and the filters. Genetic operators can then be applied to change the connections between the elements in the filters, and consequently to change the coefficient sensitivities of those filters. Fitness measure that includes the coefficient sensitivities enables the selection operator to choose low sensitivity filters. A numerical example is presented to illustrate that the sensitivity of the filter, synthesized by GP, is lower than that of the minimum coefficient sensitivity state-space realization.