A Comparison of Genetic Programming with Genetic Algorithms for Wire Antenna Design

This work compares the performance of geneticprogramming (GP) against traditional fixed-lengthgenome GA approaches on the optimization of wireantenna designs. We describe the implementation ofa GP electromagnetic optimization system for wirestructures. The results are compared with the traditionalGA approach. Although the dimensionalityof the search space is much higher for GP than GA,we find that the GP approach gives better resultsthan GA for the same computational effort. In addition,we find that a more expressive antenna structuregrammar, dramatically, improves the performance ofthe GP approach