Design and optimization of RF MEMS switch for reconfigurable antenna using feed-forward back-propagation ANN method

This paper gives the construction and parametric analysis of low loss RF MEMS switch at milli-meter and microwave frequencies up to 10GHz which support reconfigurable telecommunication devices. The switch used here exhibit excellent RF characteristics in terms of power and current consumption with typical pull-in voltages lies between 22.812 to 23.125V and switching speed of micro-second are measured which is well suited for reconfigurable microstrip antenna. The switch also result low insertion loss in the range of (0.27 to 0.17dB at DC - 3.0GHz and 0.16 to 0.06dB at 3-10GHz) in on condition and with very good isolation (144 to 141dB at DC - 3.0GHz and 141 to 137dB at 3.0-10.0GHz) at off position. Further the return loss in up and down states are calculated as -40dB to -32dB and less than -0.16dB at DC -10GHz respectively. A feed-forward back-propagation algorithm of ANN is also presented for the validation of changing the anchor arm length, width of upper beam and dielectric of RF MEMS switch. The algorithm takes these new data samples during training from finite element method (FEM) based electromagnetic simulation tool Ansoft-HFSS. Results of the neural network algorithm are compared with those of the electromagnetic (EM) solver. After optimization the particular designed switch is applied on patch antenna and simulated for generating multiband with different on- and off-configuration of switches.