High-Stability Ionic Conductive Filtering Transistors for Bio-Inspired Signal Processing.

Abstract

Applying low-pass filters in satellite communications effectively eliminates unwanted high-frequency signals and noise during image capture, mirroring the human brain's selective filtering of sensory stimuli. To enhance the efficiency of signal processing for remote sensing images, a neuromorphic information processing array based on oxide field-effect transistors are developed with HfO₂-lithium aluminium germanium phosphate (LAGP)-HfO₂ stacked dielectric (HLH FETs). The Li-ion solid-state electrolytes are stabilized in complex environments (extreme temperature and magnetic field) due to the protective sandwich structure. Meanwhile, the excellent insulating properties and Li-ion isolation effect of the high-k dielectric layer ensure a long-term reliable neuromorphic response for low-pass filtering (over one year in air). Hardware modules derived from HLH FETs are not only applicable to image processing but also show promising potential in edge computing and artificial intelligence, facilitating pattern recognition and noise reduction through biomimetic low-pass filtering functions. This innovative approach offers a new solution for modern satellite remote sensing technology and signal processing.