An Application of Genetic Programming to Electronic Design Automation: from Frequency Specifications to VHDL Code

An evolutionary algorithm is used to design a digital filter with reduced power consumption. The proposed design approach combines genetic optimisation and simulation methodology, to evaluate a multi-objective fitness function which includes both the suitability of the filter transfer function and the transition activity of digital blocks. A client-server computer program allows the user to exploit several CPU’s with parallel evolving populations, aiming at the same target specifications but with different fitness functions. Examples on design cases are presented and discussed.