Microwave MEMS Antenna Sensor Characterization and Target Detection Using Artificial Neural Networks

This paper demonstrates microwave antenna sensors fabricated on a silicon micro electro-mechanical-system (MEMS) platform. The antennas are designed to operate between 10-16 GHz with 41% bandwidth and 50 Ω input impedance. Each antenna sensor is comprised of two planar metallic four point leaves in horizontal and vertical directions providing dual linear polarization. The design is simulated using the Ansoft high frequency structure simulator. The antennas are fed using coplanar waveguide routed on top of 100 μm width silicon hinges. The fabricated MEMS platform show maximum rotation of 9.14° (total range of 18.28°) around two axes using four torsion hinges. The results highlight the limitation of silicon as hinge material when larger angles are desired. The antenna sensors demonstrate good detection of targets using measured scattered data and the artificial neural networks technique.