| Project/Area Number |
11650355
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
電子デバイス・機器工学
|
|---|
| Research Institution | Nihon University |
|---|
Principal Investigator |
SEKINE Yoshifumi Nihon University, College of Science & Technology, Professor, 理工学部, 教授 (90059965)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TAKAHASHI Sei Nihon University, College of Science & Technology, Research Associate, 理工学部, 助手 (10256810)
SAEKI Katsutoshi Nihon University, College of Science & Technology, Research Associate, 理工学部, 助手 (60256807)
|
|---|
| Project Period (FY) |
1999 – 2001
|
| Project Status |
Completed (Fiscal Year 2001)
|
| Budget Amount *help |
¥2,700,000 (Direct Cost: ¥2,700,000)
Fiscal Year 2001: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2000: ¥800,000 (Direct Cost: ¥800,000)
Fiscal Year 1999: ¥1,000,000 (Direct Cost: ¥1,000,000)
|
|---|
| Keywords | Neuron Model / Neural Network / Neural Coding / Analog Circuit / Pulse-type / Axon Model / Lambda-type Neuron Model / Chaos Neuron Model / 樹状突起モデル / Λ形ニューロンモデル / 時系列パターン識別 / ネオコグニトロン / 八木型ニューロンモデル / シミュレーション / カオス |
|---|
| Research Abstract |
We have been studying mechanisms of neural coding, and we have been trying to produce hardware from the viewpoint that development of new hardware neuron devices is one of the important problems in the study of neural networks. Furthermore, we have been studying how to develop a hardware neural network for information processing systems. In this study, we discuss as follows :
1. Investigate what happens neural coding in a nervous system.
2. Develop new neuron devices.
3. Develop temporal pattern recognition for neural network.
Results, 1. (1) It was shown that the asynchronous chaotic neuron model with effects on membrane potential on post synaptic potential had absolute refractoriness (References No. 1).
(2) The axon's output spike train displayed chaotic features when the chaotic spike train was transmitted by propagation along the active axon. Moreover, the time series intervals obtained from the axon's output spike train were an almost random train of the inter spike intervals (References No. 1).
2. (1) It was shown that the axon model, which has a certain threshold with respect to the signal height, exhibits the all-or-none law in the same way as the traditional active line, and also displays chaotic phenomena (References No. 3).
(2) We developed a pulse-type hardware bursting neuron device (References No. 4).
(3) We constructed pulse-type hardware neuron devices for neural networks (References No. 6, 7).
(4) We developed a pulse-type hardware bursting neuron device for IC implementation (References No. 9).
(5) We realized an asynchronous chaotic neuron device using analog circuits (References No. 8).
3. It was shown that by using a layered neural network, which has a structure suitable for a temporal nature, we investigated a new method of discriminating temporal patterns, such as EEG of a mouse. When we developed this neural network, we will be able to construct simple circuits, because there are not many connected lines among the neurons (References No. 5).
|
