Piecewise neural model based on split signals for Bernoulli memristors

Actuality of the investigation theme is specified by complexity of mathematical modeling of nonlinear dynamic devices, since the analytical solutions of the nonlinear differential equation systems of high size are not always obtained, and numerical solutions are often accompanied by the problem of poor conditionality. In this situation, behavioral modeling is effective, herewith the object of investigation is represented as a “black or gray box”, and its mathematical model is constructed using the sets of the input and output signals. Behavioral modeling is important in conditions of restricted information of new elements and technologies, as well as under the complexity and variety of models built at the component level. The behavioral modeling of memristive devices actively developed using nanotechnology for energy-saving equipment is represented. A method of behavioral modeling of the transfer characteristics of memristive devices by means of piecewise neural models based on split signals is proposed. To reduce the dimension on approximating nonlinear operators and, therefore, to simplify mathematical models, are applied the following: neural networks, the signal splitting method that enables to adapt the model to the type of the input signals, and a piecewise approximation method for operators of nonlinear dynamic systems. On the basis of the proposed method, a piecewise neural model is constructed. This model includes five three-layer neural networks of simple structure (3x2x1, 100 parameters) and provides a significantly higher accuracy of modeling the transfer characteristic of memristors, the current dynamics of which are described by the Bernoulli differential equation, in comparison with the two-layer piecewise neural and piecewise polynomial models. The described results are of practical value for the behavioral modeling of memristors and various memristive devices, as well as of other nonlinear dynamic systems, since they develop a universal approach for approximating nonlinear operators based on neural networks.

1. Schoukens J., Ljung L. Nonlinear system identification. A user-oriented roadmap. IEEE Control Systems Magazine. 2019;6(39):28-99. https://arxiv.org/abs/1902.00683 (accessed 05.04.2020).

2. Rogers T.J., Holmes G.R., Cross E.J., Worden K. On a grey box modelling framework for nonlinear system identification. Special topics in structural dynamics. 2017;(6):167-178. DOI: 10.1007/978-3-319-53841-9_15.

3. Bychkov Yu.A., Solovieva E.B., Shcherbakov S.V. Continuous and discrete nonlinear models of dynamic systems. Doe. 2018: 420.

4. Chua L. Memristor – the missing circuit element. IEEE Transactions on Circuit Theory. 1971;5(18):507-519. DOI: 10.1109/TCT.1971.1083337.

5. Strukov D.B., Snider G.S., Stewart D.R., Williams R.S. The missing memristor found. Nature. 2008; 7191(453):80–83. doi:10.1038/nature06932.

6. Weinstein M.Z. Electrochemical components of neuromorphic networks. The Caucasus. 2016; 3 (13): 4-11.

7. Fatima M., Begum R. Power dissipation analysis of memristor for low power integrated circuit applications. International Journal of Scientific Research in Science, Engineering and Technology IJSRSET. 2018;8(4):447-452.

8. Vourkas I., Sirkoulis G.Ch. Memristor-based nanoelectronic computing circuits and architectures. Cham, Springer International Publishing Switzerland. 2016;(19):241. DOI: 10.1007/978-3-319-22647-7.

9. Xia Q.,  Yang J.J. Memristive crossbar arrays for brain-inspired computing. Nature Materials. 2019;4(18):309-323. DOI:10.1038/s41563-019-0291-x.

10. James A. Memristor and Memristive Neural Networks. BoD–Books on Demand. 2018. DOI: 10.5772/66539.

11. Tarkov M.S. Neurocomputer systems. Internet Un-t Inform. Technology. 2016:171.

12. Khaikin S. Neural networks: a complete course. Williams Publishing House. 2019:1104.

13. Solovyeva E. Behavioural nonlinear system models specified by various types of neural networks. Journal of Physics: Conference Series (JPCS). International Conference on Information Technologies in Business and Industry. 2018;3(1015):1-6. DOI: 10.1088/1742-6596/1015/3/032139.

14. Lanne A.A. Nonlinear dynamic systems: synthesis, optimization, identification. VAS. 1985: 240.

15. Lanne A.A. Neural circuits, Hilbert's thirteenth problem and processing problems signals. Bulletin of young scientists. Technical science. 2001;7:3-26.

16. Solovyeva E. A split signal polynomial as a model of an impulse noise filter for speech signal recovery. Journal of Physics: Conference Series. International Conference on Information Technologies in Business and Industry. 2016;1(803):1-6. DOI: 10.1088/1742- 6596/803/1/012156.

17. Biolek Z., Biolek D., Biolkova V. Differential equations of ideal memristors. Radioengineering. 2015;2(24): 369-377. DOI:10.13164/re.2015.0369.

18. Georgiou P.S., Barahona M., Yaliraki S.N., Drakakis E.M. Device properties of Bernoulli memristors. Proceedings of the IEEE. 2012;6(100):1938-1950. DOI: 10.1109/JPROC.2011.2164889.

19. Ma C., Xie S., Jia Y., Lin G. Macromodeling of the memristor using piecewise Volterra series. Microelectronics Journal. 2014;3(45): 325-329. DOI: 10.1016/j.mejo.2013.11.017.

20. Solovyeva E.B., Toepfer H., Harchuk H. Behavioural model of memristors used as elements of neuromorphic systems. AIP Conference Proceedings. XIV Russian-German Conference on Biomedical Engineering. 2019;1(2140):1-4. DOI: 10.1063/1.5122000.