학술논문

Investigation of electrical characteristics of a novel FeFET-based relaxation oscillator
Document Type
Conference
Source
2023 IEEE 15th International Conference on ASIC (ASICON) ASIC (ASICON), 2023 IEEE 15th International Conference on. :1-3 Oct, 2023
Subject
Components, Circuits, Devices and Systems
Analytical models
Computational modeling
Logic gates
Threshold voltage
Iron
Oscillators
FeFETs
FeFET
relaxation oscillator
neuromorphic computing
analytical model
Language
ISSN
2162-755X
Abstract
Artificial neuron (i.e. relaxation oscillator) is one of the basic and core blocks of neuromorphic computing chips. FeFET-based relaxation oscillator has the advantages of low power consumption and compatibility with mainstream CMOS process. In this work, a closed-form analytical I-V model is developed for long channel double-gate FeFETs. Next, the oscillation dynamics of a novel oscillator with 1F1R1C structure are researched based on the proposed model. It is demonstrated that the thickness of ferroelectric (FE) film and gate voltage of FeFET have big impact on the oscillation frequency of relaxation oscillators.