Dynamical Analysis and Modeling of Experimental Data from Nonlinear Circuit Elements ("DYNAMODE")
Final Report Abstract
In this project, we performed a comprehensive circuit and system theoretical analysis focusing on experimentally verified memristor models, specifically the volatile NbOx device, the nonvolatile TaOx device, and the locally active ferromagnetic inductor all fabricated at HP Labs. We extensively analyzed their mathematical models in order to unveil the rich dynamics endowed by these devices and to reveal possible applications promoting the use of them. In practice, we focused on simulation efficiency, DC and AC characterization and local activity analysis for the NbOx memristor while we further generalized our circuit theoretic findings for other locally active volatile devices existing in the literature. As a promising application example, we designed a Reaction-Diffusion Memristor Cellular Nonlinear Network (RD-MCNN) employing NbOx elements and presented emergence of various dynamic patterns across the network. The task on modeling, analysis, and experimental realization of the ferromagnetic inductor was conducted as a master thesis work. The inductor element was built experimentally and in order to match the experimental data extracted, a self-heating based 2nd order model with an additional dynamic hysteresis loop function was defined. It was shown that the developed model was capable of replicating the experimental data successfully. Regarding the nonvolatile TaOx device, we initially reformulated the original model after resolving the numerical stability issue, included the missing window functions and unified the SET and RESET dynamics. Then, we investigated high frequency response of the same device with a focus on the fading memory property, resulting in the recognition of a novel conductance tuning mechanism for nonvolatile devices. Inspired by the exciting results, we generalized our system theoretic approach and developed a new system theoretical tool called State Change per Cycle Map (SCPCM) which generalizes multi-stability and tunability properties of nonvolatile memristors under periodic stimuli. Similarly, we applied the same theory to the volatile NbOx device and observed multistability behavior under a high frequency periodic input. In another task, we defined a physical circuit based locally active device model which can be utilized in modelling locally active dynamics. As a novelty, we designed a simple chaotic oscillator circuit experimentally implementing the new developed model through the physical circuit realization. As a final task, relying on our system-theoretic findings, we started to investigate the dynamics of a locally active memristor based oscillatory neuron circuit driven by a sinusoidal input. We perform our analysis using Volterra Series approach which essentially helps us to investigate the generation of harmonics and subharmonics at the output of the cell. The preliminary results imply that the Volterra models obtained for the single cell can be further employed for the analysis of harmonic generation mechanisms of periodically-forced coupled neurons firing at multiple frequencies and exhibiting phase locking synchronization dynamics. Due to the Covid-19 pandemic and its consequences which occurred as an exceptional case during the project work, considering the experimental work load of the project, we replaced the tasks of WP4 in the original proposal with simulation based tasks such as the system theoretical investigation of TaOx and NbOx devices. On the other hand, evaluating the outstanding results obtained, we can confidently say that the replacement of WP4 tasks can be considered as more than a compensation. The results achieved during the project work have already been published in highly valued journals and conferences and presented in several workshops while the very recent results are already submitted or yet under preparation. In general, all the project tasks were completed successfully and the outcomes of the project work have triggered new research topics which can constitute the base of upcoming research projects.
Publications
-
A Simplified Model for a NbO2 Mott Memristor Physical Realization. 2020 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE.
Messaris, I.; Tetzlaff, R.; Ascoli, A.; Williams, R. S.; Kumar, S. & Chua, L.
-
Analytical Investigation of Pattern Formation in an M-CNN with Locally Active NbOx Memristors. 2021 IEEE International Symposium on Circuits and Systems (ISCAS), 1-5. IEEE.
Demirkol, Ahmet Samil; Ascoli, Alon; Messaris, Ioannis & Tetzlaff, Ronald
-
NbO2-Mott Memristor: A Circuit- Theoretic Investigation. IEEE Transactions on Circuits and Systems I: Regular Papers, 68(12), 4979-4992.
Messaris, Ioannis; Brown, Timothy D.; Demirkol, Ahmet S.; Ascoli, Alon; Al Chawa, M. Moner; Williams, R. Stanley; Tetzlaff, Ronald & Chua, Leon O.
-
Pattern Formation in a RD-MCNN with Locally Active Memristors. Memristor - An Emerging Device for Post-Moore’s Computing and Applications. IntechOpen.
Samil, Demirkol Ahmet; Ascoli, Alon; Messaris, Ioannis & Tetzlaff, Ronald
-
A Compact and Continuous Reformulation of the Strachan TaOx Memristor Model With Improved Numerical Stability. IEEE Transactions on Circuits and Systems I: Regular Papers, 69(3), 1266-1277.
Demirkol, Ahmet S.; Ascoli, Alon; Messaris, Ioannis; Al Chawa, Mohamad Moner; Tetzlaff, Ronald & Chua, Leon O.
-
A Locally Active Device Model Based on a Minimal 2T1R Circuit. 2022 29th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 1-4. IEEE.
Demirkol, A. S.; Al Chawa, M. M.; Ascoli, A.; Tetzlaff, R.; Bedau, D. & Grobis, M.
-
Pattern Formation in an M-CNN Structure Utilizing a Locally Active NbOx Memristor. Memristor Computing Systems, 79-101. Springer International Publishing.
Demirkol, Ahmet Samil; Messaris, Ioannis; Ascoli, Alon & Tetzlaff, Ronald
-
DC Characterization of Numerically Efficient and Stable Locally Active Device Models. 2023 12th International Conference on Modern Circuits and Systems Technologies (MOCAST), 1-4. IEEE.
Demirkol, Ahmet Samil; Messaris, Ioannis; Ascoli, Alon & Tetzlaff, Ronald
-
High Frequency Response of Non-Volatile Memristors. IEEE Transactions on Circuits and Systems I: Regular Papers, 70(2), 566-578.
Messaris, Ioannis; Ascoli, Alon; Demirkol, Ahmet S. & Tetzlaff, Ronald
-
The State Change Per Cycle Map: a novel system-theoretic analysis tool for periodically-driven ReRAM cells. Frontiers in Electronic Materials, 3.
Ascoli, A.; Schmitt, N.; Messaris, I.; Demirkol, A. S.; Tetzlaff, R. & Chua, L. O.
