(Keynote) Machine Learning and High-Speed Circuitry in Thin Film Transistors for Sensor Interfacing in Hybrid Large-Area Electronic Systems. (3rd July 2019)

• Record Type: Journal Article
• Authors: Sturm, James; Mehlman, Yoni; Aygun, Levent E.; Wu, Can; Zheng, Z; Kumar, P; Wagner, Sigurd; Verma, Naveen
• Is Part Of: ECS transactions. Volume 92:Number 4(2019)
• Page Start: 121
• Page End: 134
• Publication Date: 2019-07-03
• Subjects: Electrochemistry -- Periodicals; Electrochemistry; Periodicals; Electronic journals; Electronic journal; 541.37
• DOI: 10.1149/09204.0121ecst
• Languages: English
• ISSNs: 1938-5862

Abstract

Abstract : The advent of flexible substrates with thin film transistors (TFTs) over large areas (meters) makes large-area electronics (LAE) an attractive platform for integrating very large numbers of sensors onto surfaces over large areas. While TFT's may directly interface to sensors and may be used for sensor addressing, to realistically communicate with the outside world, IC's will probably be bonded onto the "sensor sheets" to create a "hybrid" LAE/IC system. This paper examines novel architectures to minimize the number of physical interfaces to the IC, beyond the typical TFT-based active-matrix approach. Approaches demonstrated include (i) high-frequency TFT-based analog oscillators, and (ii) implementing elements of machine learning into TFT circuitry, so a higher-level information is sent to the IC's, thus requiring fewer physical connections.