Charge-trap synaptic device with polycrystalline silicon channel for low power in-memory computing

Min-Kyu Park, Joon Hwang, Soomin Kim, Wonjun Shin, Wonbo Shim, Jong-Ho Bae, Jong-Ho Lee, Seongjae Cho · 2024

Read Charge-trap synaptic device with polycrystalline silicon channel for low power in-memory computing by Min-Kyu Park, Joon Hwang, Soomin Kim, Wonjun Shin, Wonbo Shim, Jong-Ho Bae, Jong-Ho Lee, Seongjae Cho (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11585655.txt.

Description

Min-Kyu Park and co-authors build a synaptic device for computing inside memory. Their 2024 study deals with a bottleneck in current computers, where data must move between memory and processor. Processing in memory aims to replace that pattern. The team used a charge trap flash device as the synapse, the link that stores a weight. A circuit built on it performed well in tests.

The circuit cut inference energy in the synapse array, the authors report. A standard image network was trained on a common data set of small images to check accuracy. Energy, computing efficiency, and delay were all closely studied. Simulations varied device variation, array size, and chip scale. Other hardware factors were also built in. The work targets low power neuromorphic use.

Catalog record

CREATORMin-Kyu Park, Joon Hwang, Soomin Kim, Wonjun Shin, Wonbo Shim, Jong-Ho Bae, Jong-Ho Lee, Seongjae Cho
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE30 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC11585655

Browse the full library · About the Tower of Babel