[F-6] Charge-Based Memory Devices for AI Applications

[F-6-06] Junctionless Based Charge Trapping Memory for Neuromorphic Applications

〇Md. Hasan Ansari¹, Nazek El Atab¹ (1. King Abdullah University of Science and Technology (Saudi Arabia))

Presentation style: Online

https://doi.org/10.7567/SSDM.2022.F-6-06

In this work, a double gate Junctionless transistor with independent gate operations mimics human behaviors. The front gate with charge trapping in the nitride layer operates as a non-volatile memory, capturing the long-term potentiation (LTP) and depression (LTD). The back gate with single oxide operates as a floating body memory and captures the short-term potentiation (STP) of human behavior. Furthermore, the estimated conductance values are utilized for deep neural networks for image recognition where they are shown achieve 95.37% accuracy for (MNIST) pattern recognition task.