不同趋势下两种熵测度的稳定性  

Stability of entropy analysis for signal with different trends

作　　者：张印[1] 李锦[1] 姚沁[1] 赵佳[1] 王俊[2] 

机构地区：[1]陕西师范大学物理学与信息技术学院,陕西西安710119 [2]南京邮电大学图像处理与图像通信江苏省重点实验室,江苏南京210003

出　　处：《陕西师范大学学报（自然科学版）》2017年第3期29-35,共7页Journal of Shaanxi Normal University：Natural Science Edition

基　　金：国家自然科学基金(61102094)

摘　　要：通过对Logistic映射序列和采集的针灸刺激HRV(heart rate variability)信号叠加不同的趋势,探究基本尺度熵和近似熵的稳定性。结果表明:对于具有不同趋势的Logistic映射序列和心率变异性信号(针灸刺激信号),近似熵非常敏感,不能准确区别趋势中的不同信号,不适用于分析具有趋势的信号;而基本尺度熵在不同趋势下表现出较好的稳定性和准确性,该方法可广泛应用于实际信号的分析。The stability of two entropy measures, the base-scale entropy and approximate entro- py, are investigated by embedding different trends into the Logistic series and acupuncturestimu- lation heart rate variability signals. The results indicate that approximate entropy is more sensi- tive to the trends of Logistic series and acupuncture stimulation heart rate variability signals and is not suitable to analyze signal with trends. However, the base scale entropy has preferable sta- bility and accuracy for signal with different trends, which provide an experimental basis for the wide applications of the method to analyze the actual signals.

关 键 词：趋势 稳定性 基本尺度熵 近似熵 心率变异性信号 

分 类 号：TN911.7[电子电信—通信与信息系统]