论文部分内容阅读
分析了利用光折变二波耦合实现全光学输入—输出神经元响应特性,给出以光强二值化为基础的光折变单极神经元实现的可能性。分析了这种神经元在二维神经网络模型中的应用,并且利用SBN∶Cu晶体对光学神经元进行了实验模拟。
The response characteristics of all-optical input-output neurons using photorefractive two-wave coupling are analyzed, and the possibility of realizing the photorefractive monopole neuron based on binarization of light intensity is given. The application of this neuron in two-dimensional neural network model was analyzed, and the optical neuron was simulated by SBN: Cu crystal.