Abstract

In this paper, a new synapse cell with nonvolatile SONOS semiconductor memory device is proposed and it's fundamental function electronically implemented SONOS NVSM has shown characteristics that the memory value, synaptic weights, can be increased or decreased incrementally. A novel SONOS synapse is used to read out the stored analog value. For the purpose of synapse implementation using SONOS NVSM, this work has investigated multiplying characteristics including weight updating characteristics and neuron output characteristics. It is concluded that SONOS synapse cell has good agreement for use as a synapse in artificial neural networks.

Keywords

• Synapse;
• SONOS;
• ANN;
• Weights;
• Charge Potential

• 인공신경망;
• 가중치;
• 전하포텐셜;

File

Download PDF

References

1. Hybrid Analog/Digital Pulse Firing Networks A. F. Murray
2. Proc. AIP Conference on Neural Networks for Computing An Artificial Neural Networks Integrated Circuit Based on MNOS/CCD Principles J. P. Sage(et al.)
3. IEEE Trans. Electron Dev v.42 no.1 An Excellent Weight Updating Linearity EEPROM Synapse Memory Cell for Self-Learning Neuron-MOS Neural Networks T. Shibata(et al.)
4. 응용물리 v.9 no.2 저전압 비휘발성 반도체 기억소자를 위한 scaled MONOS 구조의 트랩 연구 이상배;김선주;이성배;서광열
5. 전기전자재료학회지 v.7 no.2 SONOS EEPROM 소자에 관한 연구 서광열
6. CMOS Analog Circuit Design Allan(et al.)
7. IEEE Trans. Neural Networks Pulse-Stream VLSI Neural Networks Mixing Analog and Digital Techniques A. F. Murray(et al.)
8. Neural Networks-a comprehensive foundation Simon, Haykin
9. IEEE ICNN'93 Modular Neural Networks Hardware Implementation of Analog-Digital Mixed Operation I. S. Han