Modeling, Simulation, and Emulation of Memreactive Devices for Self-organizing Oscillator Circuits
Final Report Abstract
In addition to the well-known passive electrical components, Leon Chua postulated the memristor in 1971, a nonlinear resistor with memory. It is used in neuromorphic computing architectures, among other applications, and has gained significant interest, especially in the last decade. Besides memristors, there are memreactive components such as meminductors and memcapacitors, which have been less studied and mostly appear as parasitic effects. These lossless components have the potential to make analog computers more energyefficient and compact. However, existing modeling approaches in the literature do not consider the passivity and losslessness properties of ideal memreactive components, which are crucial for circuit-theoretical studies regarding system stability. In this project, meminductors and memcapacitors were modeled in a way that ensures passivity and losslessness. A memristive LC oscillator, known as the anticipation circuit, was extended with memreactive components to allow self-organized adjustment of the resonance frequency and quality factor, as demonstrated by a wave digital (WD) emulation of the circuit. This represents a significant improvement, as it enables the anticipation of temporally irregular patterns. Further studies focused on the coupling of spatially distributed oscillators, with a particular focus on the mechanisms of topology formation. The project's results provide important insights for the modeling, emulation, and application of memreactive components and pave the way for future energy-efficient neuromorphic computing architectures.
Publications
-
Energetically consistent modeling of passive memelements. AEU - International Journal of Electronics and Communications, 93, 19-25.
Ochs, Karlheinz & Solan, Enver
-
Solving the Longest Path Problem using a HfO2-based Wave Digital Memristor Model. 2019 IEEE 62nd International Midwest Symposium on Circuits and Systems (MWSCAS), 355-358. IEEE.
Ochs, Karlheinz; Michaelis, Dennis & Solan, Enver
-
Wave digital model of a TiN/Ti/HfO2/TiN memristor. International Journal of Numerical Modelling: Electronic Networks, Devices and Fields, 32(5).
Solan, Enver; Pérez, Eduardo; Michaelis, Dennis; Wenger, Christian & Ochs, Karlheinz
-
Anticipation of Irregular Patterns: A Wave Digital Approach. 2020 IEEE 63rd International Midwest Symposium on Circuits and Systems (MWSCAS), 313-316. IEEE.
Ochs, Karlheinz; Jenderny, Sebastian & Solan, Enver
-
Mimicking Neuroplasticity by Memristive Circuits. 2020 IEEE 63rd International Midwest Symposium on Circuits and Systems (MWSCAS), 448-451. IEEE.
Ochs, Karlheinz; Michaelis, Dennis; Jenderny, Sebastian & Kohlstedt, Hermann
-
„Beiträge zur Wellendigitalemulation adaptiver Schaltungen mit gedächtnisbehafteten Bauelementen“, Dissertation, Fakultät für Elektrotechnik und Informationstechnik, Ruhr-Universität Bochum
Enver Solan
-
„Contributions to Bio-Inspired Computing via Memristive Circuits Based on Wave Digital Emulation Methods“, Dissertation, Fakultät für Elektrotechnik und Informationstechnik, Ruhr-Universität Bochum
Dennis Michaelis
-
Towards Wave Digital Modeling of Neural Pathways Using Two-Port Coupling Networks. 2022 IEEE International Symposium on Circuits and Systems (ISCAS), 809-812. IEEE.
Ochs, Karlheinz & Beattie, Bakr Al
-
A Bio-Inspired CMOS Circuit for the Excitation and Inhibition of Neuronal Oscillators. 2023 IEEE Asia Pacific Conference on Circuits and Systems (APCCAS), 31-35. IEEE.
Kumar, Singh Muralidhar Bharath; Al Beattie, Bakr; Uhlmann, Max; Ochs, Karlheinz; Kahmen, Gerhard & Rieger, Robert
