基于突触可塑性的自适应脉冲神经网络在高斯白噪声刺激下的抗扰功能研究（英文）  

Research on Disturbance Rejection of Adaptive Spiking Neural Network Based on Synaptic Plasticity under White Gaussian Noise

作　　者：郭磊[1,2] 刘东钊 黄凤荣 于洪丽[1,2] Guo Lei;Liu Dongzhao;Huang Fengrong;Yu Hongli(State Key Laboratory of Reliability and Intelligence of Electrical Equipment Hebei University of Technology,Tianjin 300130,China;Key Laboratory of Electromagnetic Field and Electrical Apparatus Reliability of Hebei Province Hebei University of Technology,Tianjin 300130,China;School of Mechanical Engineering Hebei University of Technology,Tianjin 300130,China)

机构地区：[1]省部共建电工装备可靠性与智能化国家重点实验室(河北工业大学),天津300130 [2]河北省电磁场与电器可靠性重点实验室(河北工业大学),天津300130 [3]河北工业大学机械工程学院,天津300130

出　　处：《电工技术学报》2020年第2期225-235,共11页Transactions of China Electrotechnical Society

基　　金：supported by the National Natural Science Foundation of China(61571180,51877068)

摘　　要：电磁环境的复杂多变使得传统电磁抗扰方式的不足日益凸显。生物体在神经系统调节下具有自组织、自适应和抗扰性等优势。因此,探索一种借鉴生物体优势的新的电磁抗扰方式具有重要意义,电磁仿生防护应运而生。该文基于电磁仿生防护的理念开展了基于兴奋性和抑制性突触可塑性共同调控的脉冲神经网络的抗扰功能的研究。构建了十层前馈脉冲神经网络,对高斯白噪声干扰下的脉冲神经网络进行仿真实验,重点研究了高斯白噪声对输出层神经元放电率和膜电位间相关性的影响。实验结果表明,一定强度范围的高斯白噪声干扰,对输出层神经元的放电率影响较小,输出层神经元放电率的相对变化率较小;对输出层神经元膜电位的影响较小,膜电位间的相关性较高。即在突触可塑性机制的调控下,脉冲神经网络具有一定的抗噪声干扰的能力。该研究成果为提高电子系统在复杂电磁环境下的防护能力奠定了理论基础。With the electromagnetic environment becoming more and more complex,the shortcomings of traditional anti-electromagnetic disturbance methods are becoming increasingly prominent.The organism under the regulation of nervous system has the advantages on self-organization,self-adaption and disturbance rejection.It is significant to explore a new method of anti-electromagnetic disturbance based on the excellent characteristics of organism.Therefore,a new field of electromagnetic bionic protection emerged.In this paper,the disturbance rejection of the spiking neural network based on excitatory synaptic plasticity and inhibitory synaptic plasticity was analyzed.A ten-layer feed-forward spiking neural network was constructed.The firing rate and correlation between membrane potential of the neuron were considered as indexes for assessing disturbance rejection ability.The experimental results show that:under a certain intensity of white Gaussian noise disturbance,the relative variation of output firing rate is tiny;the correlation between membrane potential in output layer is relatively large.It is concluded that the spiking neural network based on synaptic plasticity has a certain ability to reject noise disturbance.This study lays the theoretical foundation for improving the protection ability of electronic system in the complex electromagnetic environment.

关 键 词：脉冲神经网络 突触可塑性 抗扰 高斯白噪声 

分 类 号：Q189[生物学—神经生物学] TP391.9[生物学—普通生物学]