This paper introduces a method for modeling and analyzing neural impulse sequences. In this paper, we define the response value of a scale-independent neuron and construct the correlation graph of the neuron under the response value. The minimum cut algorithm is applied continuously to obtain the maximum group of neurons. According to the characteristics of the firing of neurons, a Poisson-process based model is proposed to mathematically model the neural coding, and the gradient descent method is used to optimize it. Through the modeling analysis method, information such as maximum neuron group and Inter-spike-Interval (ISI) can be effectively analyzed according to neuron impulse sequence.

Mathematical modeling is of great significance to the study of brain function. In this paper, an analysis method based on Poisson’s general linear model is adopted to model the brain impulse sequence. First, the ideas of Hodegkin-Hulex model, STA and Poisson’s general linear model are introduced. Secondly, the implementation of the general linear model is introduced. Finally, Poisson GLM Fit Spike-train is simulated and compared with Linear Gaussian-GLM. The AIC tradeoff estimation model fits well. The superiority of Poisson GLM Fit Spike-Train is demonstrated