Parallel signal processing of wireless pressure sensing platform combined with machine learning based cognition, inspired from human somatosensory system

Abstract

Inspired from the human somatosensory system, pressure applied to multiple pressure sensors have been received in parallel and combined into a representative signal pattern, which was subsequently processed using machine learning. The pressure signals were combined using a wireless system, where each sensor was assigned a specific resonant frequency on the reflection coefficient ($S_{11}$) spectrum, and the applied pressure changed the magnitude of the S11 pole with minimal frequency shift. This allowed the differentiation and identification of the pressure applied to each sensor. The pressure sensor consisted of polypyrrole-coated microstructured PDMS placed on top of electrodes, operating as a capacitive sensor. The high dielectric constant of polypyrrole enabled relatively high pressure sensing performance. The coils were vertically stacked to enable the reader to receive the signals from all of the sensors simultaneously at a single location, analogous to the junction between neighboring primary neurons to a secondary neuron. Here, the stacking order was important to minimize the interference between the coils. Furthermore, convolutional neural network-based machine learning was utilized to predict the applied pressure of each sensor from unforeseen $S_{11}$ spectra. With increasing training, the prediction accuracy improved (with mean squared error of 0.12), analogous to humans’ cognitive learning ability.