A5029040972- Advanced Memory and Neural Computing
- Neural dynamics and brain function
- Ferroelectric and Negative Capacitance Devices
- Fault Detection and Control Systems
- Neuroscience and Neural Engineering
- stochastic dynamics and bifurcation
- Semiconductor materials and interfaces
- Cybersecurity and Information Systems
- Machine Fault Diagnosis Techniques
- Transition Metal Oxide Nanomaterials
- Photoreceptor and optogenetics research
- Geophysics and Sensor Technology
- Semiconductor materials and devices
- Anomaly Detection Techniques and Applications
- Mechanical and Optical Resonators
- Force Microscopy Techniques and Applications
- Advanced Data Processing Techniques
Scientific and Production Association of Electromechanics (Russia)
2023
Space Research Institute
2023
Research Institute of Molecular Electronics
2019-2022
Moscow Institute of Physics and Technology
2019-2021
Memristors are among the most promising devices for building neural processors and non-volatile memory. One circuit design stage involves modeling, which includes option of memristor models. The common approach is use compact models, accuracy often determined by their parameter extraction from experiment results. In this paper, a review existing methods was performed new algorithms an adaptive model were proposed. effectiveness developed confirmed volt-ampere characteristic with vertical...
The compatibility of memristor materials with advanced complementary metal-oxide-semiconductor (CMOS) technology is a key factor for microelectronics element base manufacturing. Therefore, we continued studying previously fabricated CMOS-compatible Ni/Si3N4/SiO2/p+-Si samples. We approximated volt-ampere characteristics (VAC) at different temperatures using the general form spatial charge-limiting current (SCLC) equation assuming exponential and Gaussian trap distribution within band gap...
The electromigration influence on main characteristics of planar electrochemical systems with difference parameters was studied. transporting processes are simulated, current-voltage characteristic, transfer function and THD were calculated, quantitative estimates the magnitude asymmetric flow its effect nonlinearity coefficient system obtained.
The memristor is one of the modern microelectronics key devices. Due to nanometer scale and complex processes physic, development state study approaches faces limitations classical methods observe processes. We propose a new approach investigate degradation six Ni/Si3N4/p+Si-based memristors up their failure. basis proposed idea joint analysis resistance change curves with volt-ampere characteristics registered by auxiliary signal. paper considers existence stable switching regions...
The use of low-dimensional materials is a promising approach to improve the key characteristics memristors. development process includes modeling, but question most common compact model applicability modeling device with inclusion remains open. In this paper, comparative analysis linear and nonlinear drift as well threshold models was conducted. For purpose, assumption relationship between results optimization volt–ampere characteristic loop descriptive ability used. A global random search...
The paper deals with the problem of controlling state industrial devices according to readings their sensors. current methods rely on one approach feature extraction in which prediction occurs. We proposed a technique build scalable model that combines multiple different extractor blocks. A new based sequential sensor space analysis achieves state-of-the-art results C-MAPSS benchmark for equipment remaining useful life estimation. resulting performance was validated including changes scaling.
The work is devoted to the analysis of various approaches problem empirical memristor model parameters extraction. A description peculiarities extraction process given, and an original version algorithm proposed. proposed compared with other considered ones.