Analog reservoir computing via ferroelectric mixed phase boundary transistors.
Jangsaeng Kim, Eun Chan Park, Wonjun Shin, Ryun-Han Koo, Chang-Hyeon Han, He Young Kang, Tae Gyu Yang, Youngin Goh, Kilho Lee, Daewon Ha, Suraj S Cheema, Jae Kyeong Jeong, Daewoong Kwon
Author Information
Jangsaeng Kim: Department of Electronic Engineering, Hanyang University, Seoul, 04763, Republic of Korea. ORCID
Eun Chan Park: Department of Electronic Engineering, Hanyang University, Seoul, 04763, Republic of Korea.
Wonjun Shin: Department of Electrical and Computer Engineering and Inter-university Semiconductor Research Center, Seoul National University, Seoul, 08826, Republic of Korea. ORCID
Ryun-Han Koo: Department of Electrical and Computer Engineering and Inter-university Semiconductor Research Center, Seoul National University, Seoul, 08826, Republic of Korea.
Chang-Hyeon Han: Department of Electronic Engineering, Hanyang University, Seoul, 04763, Republic of Korea.
He Young Kang: Department of Electronic Engineering, Hanyang University, Seoul, 04763, Republic of Korea.
Tae Gyu Yang: Department of Electronic Engineering, Hanyang University, Seoul, 04763, Republic of Korea.
Youngin Goh: Semiconductor Research and Development Center, Samsung Electronics, Hwaseong, Republic of Korea.
Kilho Lee: Semiconductor Research and Development Center, Samsung Electronics, Hwaseong, Republic of Korea.
Daewon Ha: Semiconductor Research and Development Center, Samsung Electronics, Hwaseong, Republic of Korea.
Suraj S Cheema: Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, MA, 02139, USA. sscheema@mit.edu. ORCID
Jae Kyeong Jeong: Department of Electronic Engineering, Hanyang University, Seoul, 04763, Republic of Korea. jkjeong1@hanyang.ac.kr. ORCID
Daewoong Kwon: Department of Electronic Engineering, Hanyang University, Seoul, 04763, Republic of Korea. dw79kwon@hanyang.ac.kr. ORCID
Analog reservoir computing (ARC) systems have attracted attention owing to their efficiency in processing temporal information. However, the distinct functionalities of the system components pose challenges for hardware implementation. Herein, we report a fully integrated ARC system that leverages material versatility of the ferroelectric-to-mixed phase boundary (MPB) hafnium zirconium oxides integrated onto indium-gallium-zinc oxide thin-film transistors (TFTs). MPB-based TFTs (MPBTFTs) with nonlinear short-term memory characteristics are utilized for physical reservoirs and artificial neuron, while nonvolatile ferroelectric TFTs mimic synaptic behavior for readout networks. Furthermore, double-gate configuration of MPBTFTs enhances reservoir state differentiation and state expansion for physical reservoir and processes both excitatory and inhibitory pulses for neuronal functionality with minimal hardware burden. The seamless integration of ARC components on a single wafer executes complex real-world time-series predictions with a low normalized root mean squared error of 0.28. The material-device co-optimization proposed in this study paves the way for the development of area- and energy-efficient ARC systems.
References
Nat Commun. 2020 May 15;11(1):2439
[PMID: 32415218]
Nat Commun. 2017 Dec 19;8(1):2204
[PMID: 29259188]
