Neuromorphic Time-Dependent Pattern Classification with Organic Electrochemical Transistor Arrays

TL;DR

This paper demonstrates a neuromorphic pattern classification system using organic electrochemical transistor arrays, leveraging ionic dynamics and reservoir computing to handle variability and enable real-time sensing and processing.

Contribution

It introduces a novel bio-inspired hardware approach with iono-electronic materials for pattern classification, addressing fabrication variability and demonstrating real-time capabilities.

Findings

Effective classification with a 12-unit array despite variability

Use of a new electropolymerizable polymer for device stability

Performance analysis of array size and variability impacts

Abstract

Based on bottom-up assembly of highly variable neural cells units, the nervous system can reach unequalled level of performances with respect to standard materials and devices used in microelectronic. Reproducing these basic concepts in hardware could potentially revolutionize materials and device engineering which are used for information processing. Here, we present an innovative approach that relies on both iono-electronic materials and intrinsic device physics to show pattern classification out of a 12-unit bio-sensing array. We use the reservoir computing and learning concept to demonstrate relevant computing based on the ionic dynamics in 400-nm channel-length organic electrochemical transistor (OECT). We show that this approach copes efficiently with the high level of variability obtained by bottom-up fabrication using a new electropolymerizable polymer, which enables iono-electronic device functionality and material stability in the electrolyte. We investigate the effect of the array size and variability on the performances for a real-time classification task paving the way to new embedded sensing and processing approaches.