Large array antennas are of great interest for 5G / 6G mobile communication systems. However, when designing and optimising the performance of such systems it is computationally expensive to use a full wave electromagnetic model of all the antenna elements at once, even though a full model of a single element is relatively easy.
Consequently, simple models of each element are often used – models that may not contain enough information to give full confidence in the design / optimisation process. An optimisation process, e.g. evolutionary optimisation, may then find an optimum design that is a false optimum of the simplified model that is not present in a real design.
At York we have over 20 years of experience in evolutionary optimisation of antenna designs. This project will build on this work to consider how to develop large-scale models of array antennas that retain enough electromagnetic information to accurately model the interactions – but, while requiring sufficiently small computational resources that allow for strong optimisation techniques to be applied.