An Empirical Study of Computational Optimisation Techniques for Microstrip Antennas

There are many computational optimisation techniques, several of which have been
applied to real world problems, such as wire antennas, building structures and turbine
blade profiles. Some of these techniques are relatively well known within the scientific
and engineering communities, such as genetic algorithms. Microstrip antennas (MSAs) are
widely used, especially for mobile communications applications, due to their low profile and
low cost. An empirical study was performed to ascertain which computational optimisation
technique is the most efficient when optimising MSAs. In this context, the most efficient
technique refers to the one that has the highest probability of finding a solution that meets
the required specification when all techniques have the same computational time allocated
to them. It was found that genetic algorithms, the simplest technique used, is the most
efficient of those that were tried. The main reason for this was concluded to be due to
the relatively low number of fitness evaluations performed per run. Other, more complex,
techniques are likely to to be more efficient when more fitness evaluations (run time) are
available.