Scientific results:
The Laboratory has walked a path from solving fundamental problems of materials science, the physics of complex nonlinear stochastic systems to memristive devices in the integrated design, new physical and mathematical models accounting for the effect of fluctuations and analog-digital circuits relying on memristors. We generalized the results of a theoretical and experimental research of fluctuation phenomena in multistable systems based on memristive materials. Our research experimentally demonstrates fundamentally new possibilities for increasing the stability, forecasting the behavior and controlling memristive devices in prototypes of electronic appliances and next-generation neuromorphic systems. The Laboratory has developed technological recommendations for creating resistive memory elements and devices based on them that demonstrate stability of behavior and the possibility of controlling the parameters. We conducted a research of the role of stochastic factors and noises in the trainability of pulse neural network architectures based on synaptic memristive structures .
On the basis of the conducted research of fluctuation phenomena in memristive systems, we have created a stochastic macromodel of a memristor that becomes an efficient tool in the development of a new element base and automated design tools for a whole range of products relying on memristors. As we expected and as the conducted research demonstrated, fluctuations can play a constructive role in memristive systems.
Implemented results of research:
The performed research allows to find applications of the discovered phenomena and implement the proposed methods of improving the characteristics of electronics based on memristive materials. This takes the project to a new applied level that requires orientation towards concrete electronic devices, products related to them, and close cooperation with organizations of the real sector of the economy.
The most obvious application of memristive devices is universal resistive RRAM (resistive random access memory), which unites the properties of random access memory and persistent memory devices, as well as possesses unique radiation resistance. This is important for applications in space and in nuclear reactors. Among the promising areas of use of memristive devices is energy-efficient hardware neural network (neuroprocessors) for communication systems, including IoT, big data processing, robotics and artificial intelligence. In the future, a new technological breakthrough can be related to the creation of adaptive neuroprosthetics and neurointerfaces on the basis of brain-like memristive networks to replace, extend or recover lost brain functions.
Education and career development:
Since 2018, we have been conducting a regular research and education seminar. The regular seminar was organized by order of the Rector of Nizhniy Novgorod State University for undergraduate and postgraduate students and young researchers. The main goal of the seminar is to present the results of the Laboratory’s research work in the domain of comprehensive research of fluctuation phenomena in multistable systems for the creation of new generations of electronic devices and neuromorphic artificial intelligence technologies based on memristive technologies as well as related domains.
Employees of the Laboratory have organized two international scientific seminars in Russia and Greece, two international conferences in Spain, one international symposium in Russia and 3 Russian scientific conferences.
In 2019, Professor Spagnolo taught a lecture course on the current problems of statistical physics for students and employees of Nizhniy Novgorod State University.
Employees of the Laboratory provide scientific supervision to undergraduate and postgraduate students of Nizhniy Novgorod State University.
In 2018–2020, 29 employees of the Laboratory have completed advanced training programs.
Organizational and structural changes:
The Satellite Oceanography Laboratory (StoLab, http://www.stolab.unn.ru/) functions in Nizhniy Novgorod State University within the Research and Education Center «Physics of Solid-state Nanostructures». The Laboratory is interdisciplinary and consists of 5 sectors: a sector of modern stochastic analysis methods, a sector of memristive material technologies, a sector for microscopic probing, a sector of the physics of noises, a sector for neuromorphic technologies. The interdisciplinary laboratory has united the efforts of leading scientific, pedagogical and engineering personnel of Nizhniy Novgorod State University (the Faculty of Radiophysics, the Research and Education Center «Physics of Solid-state Nanostructures», the Faculty of Physics, the Research Institute of Physics and Technology, the Nizhniy Novgorod Neuroscience Center) and employees of the Kurchatov NBICS-centre (Kurchatov Institute). The research team consists of 50 employees, including 5 Doctors of Sciences, 18 Candidates of Sciences, 8 undergraduate students of Nizhniy Novgorod State University and 6 postgraduate students of Nizhniy Novgorod State University.
Collaborations:
On the basis of Nizhniy Novgorod State University active collaboration is pursued with employees of a number of divisions of the Research Institute of Neurosciences and the Institute of Biology and Biomedicine aimed at designing and creating new generations of neuromorphic and neurohybrid systems based on the developed memristive schemes and living networks (cultures and tissues) of the brain to solve current problems of robotics, artificial intelligence and medicine. To develop original scientific and technological solutions in the photolithography of thin-film structures as part of memristive structures, the Laboratory collaborates with employees of the Department of Physical Chemistry of the Faculty of Chemistry of Nizhniy Novgorod State University.
Two foreign researchers from the University of Palermo (Italy), Bernardo Spagnolo and Angelo Carollo, work for our Laboratory.
The Laboratory conducted joint research with the participation, pursuant to a remote work agreement, of Sun Jun Kim, associate professor at Jeonbuk National University (South Korea). The research was aimed at designing and studying new metal-oxide memristive devices for neuromorphic systems. A joint article has been published.
The Laboratory has been conducting joint research with the participation, via a remote work agreement, of Alexander N. Pisarchik, chair of the Laboratory of Computational Biology of the Center of Biomedical Technology of Polytechnic University of Madrid (Spain). The research is aimed at developing a neurohybrid interface based on the self-organization of multistable memristive and living neural systems. A joint article has been published.
Partners of StoLab:
- Kurchatov Institute, Moscow, Russia
- University of Leicester, Great Britain
- Loughborough University, Great Britain
- Polytechnic University of Madrid, Center for Biomedical Technology, Madrid, Spain
- University of Granada, Department of Electronics and Computer Technology, Granada, Spain
- National Center for Scientific Research Demokritos, Institute of Nanoscience and Nanotechnology, Athens, Greece
- University of South Carolina, Columbia, Department of Physics and Astronomy, United States
- Tsinghua University, Institute of Microelectronics, Beijing, China
- Dongguk University, Division of Electronics and Electrical Engineering, Seoul, Republic of Korea
- Indian Institute of Technology (IIT) Ropar, Department of Electrical Engineering, Guwahati, India
- «Engineering Center of Lobachevskiy University» Ltd
- Branch of «Russian Federal Nuclear Center – All-Russian Research Institute of Experimental Physics» – Y. E. Sedakov All-Russian Scientific Institute of Measuring Systems, Nizhniy Novgorod, Russia
- JSC «Research Institute of Molecular Electronics», Zelenograd, Moscow, Russia
The collaboration with the above-mentioned partners resulted in the publication of articles in Q1 journals and chapters in collective monographs.
