Research Article
Signal Amplification by Circular Single-Electron Oscillator Network with Stochastic Resonance
@INPROCEEDINGS{10.4108/icst.bict.2014.257850, author={Hiroyuki Otake and Kazuyoshi Ishimura and Tetsuya Asai and Takahide OYA}, title={Signal Amplification by Circular Single-Electron Oscillator Network with Stochastic Resonance}, proceedings={8th International Conference on Bio-inspired Information and Communications Technologies (formerly BIONETICS)}, publisher={ICST}, proceedings_a={BICT}, year={2015}, month={2}, keywords={neural network stochastic resonance single-electron circuit}, doi={10.4108/icst.bict.2014.257850} }- Hiroyuki Otake
Kazuyoshi Ishimura
Tetsuya Asai
Takahide OYA
Year: 2015
Signal Amplification by Circular Single-Electron Oscillator Network with Stochastic Resonance
BICT
ACM
DOI: 10.4108/icst.bict.2014.257850
Abstract
In this paper, a new type of a neuromorphic single-electron(SE) circuit is described and its operation is also evaluated by conducting Monte Carlo simulation. The circuit is expected as a new information-processing device. However, it is known that the circuit is very sensitive to noise and fluctuation, so that we have studied how to overcome such noise effect. Recently, unique technique that imitates stochastic resonance (SR) behavior in neural networks has been proposed to solve the problem of noise. We focus on internal noise for the SR system. To confirm the SR phenomenon caused by the internal noise, we change the circuit form to a circular structure. We can construct the circular neural network by using the SE circuit because the operation of them is similar to neurons. The proposed SE circuit network was able to show the SR phenomenon. Moreover, the circuit we found could amplify the output signal.