Spike propagation by synchronization and vibrational resonance in feedforward Izhikevich neural network

Multilayer feed − forward neuron networks (FFNs) are the basis of various machine learnings, in which the propagation of neural firing rates with respect to synchronization and vibrational resonance (VR) under white Gaussian noise and high − frequency stimulation (HFS) is important. In this study, the influences of HFS and noise on propagation about synchronous firing rate and the VR are investigated in different kinds of Izhikevich neural FFN. For ten layers excitatory Izhikevich neuron network, it is shown that synchronous firing rates appear gradually and the diverse noise intensities, synaptic weights and time constants take effects on the propagation of synchronous discharge rates. For four layers excitatory FFN, it is observed that the systemic output rates keep smaller than zero, and carries no information of weak signal by using small HFS. The VR phenomenon occurs, when the input of weak low frequency signal (LFS) and signal output keep well phase synchronization, and the LFS is amplified by increasing amplitude of HFS. In the excitatory − inhibition neural multilayer FFN, the propagation by synchronous firing rates is not well, and few inhibition neurons are keep excited. In the systemic output, the synchronization phenomenon can be observed, but not as good as in excitatory neural FFN.