MLP Neural Model of Dual-Band Slotted Square Patch Antenna with Variable Slot Geometry

This paper presents a neural model of a slotted square patch antenna based on a MultiLayer Perceptron network (MLP), which uses the EM software “SSPA_design” to generate samples needed for training and testing of the neural model. The input parameters of the neural model are the length of patch (L), the feed position coefficient (k
f
), the slot position coefficient (k
d
), the slot length coefficient (k
a
) and the length-to-width ratio of the slot (k
ab
), while the output parameters are resonant frequencies belonging to the range [700-3000] MHz.