Name
Abstract | ||
|---|---|---|
Silicon Oxide (SiO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
)-based memristor emerges as a promising candidate for 3-D memory, logic, and neuromorphic computing applications due to their high endurance, low cost and easy fabrication in the CMOS processing line. Even though many efforts have been devoted to understanding memristive mechanisms in SiO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
, their performance benefits over the most investigated metal-oxide-based memristors are still unclear. In this regard, we investigate the performance potential of SiO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
-based memristor and explore their benefits over HfO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
and TaO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
-based memristors for nonvolatile memory and memristive systems. The performance evaluation of SiO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
and metal-oxide based memristors is done using a fully calibrated self-developed three-dimensional (3-D) simulation, based on self-consistent solutions of drift-diffusion, current continuity, and the heat equations. The SiO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
-based memristor is found to be a promising candidate over HfO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
and TaO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
-based memristors architecture for nonvolatile memory with a higher on-off current ratio (400), lower sneak current (0.43 $\mu$A), faster-switching speed (20 ns). Further, a larger conductance variation with pulse number for SiO
<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf>
-based memristor promises a superior performance with analog memory and emulating synaptic components. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1109/VLSID2022.2022.00064 | 2022 35th International Conference on VLSI Design and 2022 21st International Conference on Embedded Systems (VLSID) |
Keywords | DocType | ISSN |
Memristor,neuromorphic computing,nonvolatile memory,silicon oxide,HfOₓ,TaOₓ | Conference | 1063-9667 |
ISBN | Citations | PageRank |
978-1-6654-9661-2 | 0 | 0.34 |
References | Authors | |
2 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Mani Shankar Yadav | 1 | 0 | 0.34 |
| Avinash Kumar Gupta | 2 | 0 | 0.34 |
| Kanupriya Varshney | 3 | 0 | 0.34 |
| Brajesh Rawat | 4 | 0 | 0.34 |