基于SR-UKF的神经元群模型分析

Analysis of neuronal population model based on SR-UKF

外部输入通过神经元群模型可以产生不同类型的类似EEG的信号,但是外部输入信号的统计特性多采用经验值,而没有对其专门进行研究。在设定神经元群模型参数的情况下,SR-UKF被用来估计不同观测值所对应外部输入以及模型输出。实验证明外部输入估计数据的均值在前人所采用值的范围内,但是标准差比前面用到的小很多。

Signals similar to EEG can be produced while extrinsic input passes through the model,but empirical values are used to describe the statistical property of extrinsic input without special research on it. In the context of given neuronal population model parameters,SR-UKF is used to estimate the extrinsic input and model output corresponding to different observations. The results show that the mean values of estimated extrinsic inputs are within the interval recommended by previous literatures,but the standard deviations are far less than the experience value used before.