Design of Variation-Tolerant 1F-1T Memory Array for Neuromorphic Computing

• DOI: 10.36227/techrxiv.22178192.v1
• Publication Date: 2023-03-02T02:20:48Z
• Authors: Masud S K Rana; Sunanda Thunder; Franz Müller; Nellie Laleni; Yannick Raffel; Maximilian Lederer; Luca Pirro; Talha Chohan; Jing-Hua Hsuen; Tian-Li Wu; Konrad Seidel; Thomas Kaempfe; Sourav De; Bhaswar Chakrabarti
• Keywords: Neuromorphic engineering; MNIST database; Perceptron; Limiting; Non-Volatile Memory; Array data structure

<p>This letter proposes a memory cell, denoted by 1F-1T, consisting of ferroelectric field-effect transistor (Fe-FET) cascoded with another current-limiting (T). The reduces the impact drain current (Id) variations limiting on-state in FeFET, 1F. We have fabricated 28nm high-k-meta-gate (HKMG) based FeFETs, and experimental data is used to model simulate single-cell arrays. simulation shows significant improvement bit-line (BL) (IBL ) variation for 1F-1T array. Finally, system-level neuromorphic synapses an inference accuracy 97.6% MNIST hand-written digits multi-layer perceptron (MLP) neural networks.</p>