非高斯分布噪声下诱发电位潜伏期变化自适应检测

Detection of adaptive latency change of evoked potentials under Gaussian noise conditions

传统的高斯分布白噪声的模型不能很好地描述 EP信号中脑电图 (EEG)和其他噪声的特性 .因此 ,根据α稳定分布噪声理论和 EP信号中噪声的非高斯特性 ,提出了一种基于最小分数低阶矩的自适应诱发电位潜伏期估计方法 .这种方法既可以应用于高斯噪声环境 ,又在低阶α稳定分布噪声 (一类典型的非高斯噪声 )环境下具有良好的韧性 ,是一种可靠的检测EP信号潜伏期变化的方法 .分析和实验表明 ,α稳定分布噪声模型是一种适合于描述带噪EP信号统计特性的随机噪声模型 ,所得到的 EP信号潜伏期变化的检测结果 。

Traditional evoked potential (EP) analyses are developed under the condition that the background noises are Gaussian distribution. However, Gaussian distribution doesn′t describe the EP signals and the contaminated noises (such as EEG and other noises) properly. Based on the α \|stable distribution theory and the properties of EP signals and noises, this paper proposes a latency change detection and estimation algorithm under α \|stable noise conditions, a generalization of Gaussian noise assumption. The simulation and the analysis show that the α \|stable model fits the noises found in the impact acceleration experiment under study better than the Gaussian model. The proposed algorithm is more robust under both Gaussian and lower order α \|stable noise conditions.