欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
一些研究表明,脑电信号具有低维的混沌动力学特性,脑电的维数则反映脑信息处理过程中的神经元集群的活动状态,是研究脑电信号的重要非线性参数。在Grassberger-Procaccia计算关联维数算法的基础上,通过改进计算关联积分的过程,提出一种适合于移动重叠窗口的维数复杂度连续检测方法。另外,还对关联积分线性区间选择上做了改进,使得所求得的参数在严格意义上为维数复杂度。数值仿真验证了该方法的正确性,并在此基础上对视觉输入变化的EEG信号进行动态数据分析,结果显示睁眼时脑电的复杂度明显高于闭眼时脑电的复杂度。所提出的维数复杂度计算方法减少了数据分析的负荷,适合于连续数据分析,能够准确地反映脑电维数复杂度的连续变化过程。
英文摘要：
It is indicated that EEG reflects dynamics on low- dimensional chaotic attractors. Chaotic dimension of EEG is an important nonlinear parameter reflecting the activities of neuron populations in brain when processing information. Based on Grassberger- Procaccia method of correlation dimension, we proposed an improved method for detecting dimension complexity in processing overlapped window data. In addition, scaling region of correlation integral was also chosen by a new method. The calculated value represented dimension complexity instead of correlation dimension. Numerical simulation verified the availableness of the method, and the analysis of EEG signal related to visual input indicated that dimension complexity during eye opening was greater than that during eye closing. The proposed method decreased the overload of calculation and adapts for continuous data analysis and could reflect the continuous variation of EEG dimension complexity as well.
同期刊论文项目
同项目期刊论文
期刊信息
位置：