In this paper, the design of frequency reconfigurable planar antenna by incorporation of metasurface superstrate (FRPA-MSS) is presented using an artificial neural network. The dual-layer radiating structure is created on a 1.524 mm thick Rogers RO4350B substrate board (εr = 3.48, tan δ = 0.0037). The candidate antenna is designed and analyzed using a high-frequency structure simulator (HFSS) tool. The transfer matrix method is employed for the successful retrieval of electromagnetic properties of the metamaterial. Frequency reconfiguration is achieved by placing the metasurface superstrate onto the rectangular patch antenna. A simplified ANN approach has been employed for the design of metasurface incorporated proposed antenna. Presented prototypes are characterized through experimental measurements. It is found from the practical observations that the proposed antenna effectively reconfigures the tuning range from 5.03 to 6.13 GHz. Moreover, the presented antenna operates efficiently with agreeable gain, good impedance matching, and stable pattern characteristics across the entire operational bandwidth. The experimental results obtained validate the simulated performance.

Coupled-mode theory is applied to obtain an analytic form of scattering parameters for a class of transmission line metamaterials with antisymmetric split-rings. The same structure is modeled with equivalent circuit, which includes electric and magnetic coupling with the line and inter-resonator coupling. Modified even/odd analysis is used to obtain scattering parameters from the equivalent circuit. These two methods are shown to be equivalent in a narrow band, and their constants related. The obtained results are compared with full-wave simulations and measurements, and it is shown that both methods give accurate approximation in one octave frequency band. The derived analytic expressions are suitable for study of resonant phenomena, with potential practical applications for filters, phase shifters, delay lines, and sensors.