摘要
针对地面环境监控中的人员脚步信号识别问题,提出一种基于希尔伯特-黄变换的识别算法。利用经验模态分解方法将探测得到的震动信号分解成若干个固有模态函数分量,找出各分量在频谱上的奇异点,通过对这些奇异点的希尔伯特谱进行判定,实现对人员脚步信号的识别。实验结果证明,该算法具有较高的正确识别率。
Aiming at the human footstep signal identification of the ground environment monitoring,an identification algorithm based on Hilbert-Huang Transform(HHT) is proposed.By using Empirical Mode Decomposition(EMD),the vibrating signal is decomposed into a collection of Intrinsic Mode Function(IMF) and find out whether there is singular point or not on the spectrum of these IMF.It judges the singular point on the basis of the Hilbert spectrum and thus makes out the human footstep.Experimental results show that the algorithm has high accuracy.
出处
《计算机工程》
CAS
CSCD
北大核心
2011年第19期168-170,173,共4页
Computer Engineering
基金
国家"973"计划基金资助项目(2011CB302906)
国家科技重大专项基金资助项目(2010ZX03006-004)
关键词
地面环境监控
人员脚步信号识别
希尔伯特-黄变换
震动信号
希尔伯特谱
ground environment monitoring
human footstep signal identification
Hilbert-Huang Transform(HHT)
vibrating signal
Hilbert spectrum