Novel Modeling Methodology for Memristive Devices for Circuit Design and Simulation
This work introduces a methodology for obtaining the high-level model of memristive devices. The methodology uses the I- V curve points of the device. Model equations are derived from the calculation of a state-variable of a first-order differential equation that describes the dynamics of the device. As a demostration, the procedure has been applied to three classical memristor models from which new modeling equations have been developed. Comparison between the resulting model and the initial data points is carried out by measuring the Mean Absolute Percentage Error (MAPE), resulting in 4%, 2.8% and 20.1% when compared to original Strukov's, Biolek's and Chang's models, respectively. This methodology simplifies the simulation process by maintaining a first-order state variable for the model, and avoiding highly non-linear equations such as those introduced by window funcions in the port equation.
Authors
- Arturo Sarmiento-Reyes (ORCID: https://orcid.org/0000-0002-7619-4994)
- Jesús Jiménez-León (ORCID: https://orcid.org/0000-0003-0654-1775)
- Pedro Rosales (ORCID: https://orcid.org/0000-0002-8352-5114)
Institutions
- National Institute of Astrophysics, Optics and Electronics (MX)
Publication Details
- Published
- 2024-10-23
- DOI
- https://doi.org/10.1109/cce62852.2024.10771006
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00