Construction And Simulation Study Of Spiking Neural Network Based On Synaptic Plasticity

As we all know,the brain may be the Earth or the world's most sophisticated machines,hundreds of millions of neurons and the connection between them determines the extraordinary characteristics of the brain,neuroscience research shows that the biological signal generation and transmission of Information encoding and transmission mechanism has a very important role.The network topology structure and the connections between neurons are the basis of neural network modeling.External noise stimulation and synaptic weight distribution are also important factors affecting the neuronal firing.In addition,the number and type of network nodes And synaptic plasticity mechanisms also affect the network discharge characteristics.Based on the model of spiking neuron network,the effect of synaptic plasticity on excitatory and inhibitory neurons on the synchronization characteristics of neurons was analyzed for weak signal transmission and neuronal firing.(1)Neuron model selection,contrasted various neuronal models,selected Izhikevich neuronal model.The synaptic plasticity was analyzed and the spiking neural network with excitatory synaptic plasticity was constructed and its discharge characteristics were analyzed.It is proved that there is a certain correlation between the connection probability and the input information.The low-link probability network can transmit the information better.(2)Compared with excitatory neurons and inhibitory neurons,the spiking neural networks with excitatory and inhibitory plasticity were constructed and their discharge characteristics were studied.It was confirmed that the excitatory and inhibitory neurons and the synaptic plasticity between them are closely related to the network's adaptive synchronous discharge characteristics.It is found that excitatory neurons promote the release of postsynaptic action potentials under STDP mechanism,whereas inhibitory neurons decrease the rate of postsynaptic action potentials under synaptic plasticity regulation.(3)The spiking neural network with complete synaptic plasticity was subjected to noise excitation and its discharge characteristics were analyzed.It is found that the frequency oscillation with the middle layer of the spiking neural network becomes lower and the discharge frequency tends to be consistent in noisy environment.The spiking neural network with complete synaptic plasticity was subjected to different intensity noise excitations and the discharge characteristics were analyzed.It was proved that the effect of noise intensity on frequency was obvious,and the influence of connection mode on network frequency synchronization could be weakened to a certain extent.