摘要
针对固定时间窗函数短时傅里叶变换(STFT)时频表示时间分辨率和频率分辨率相互制约,影响时频聚集性的缺陷,提出了一种组合窗函数的STFT时频表示方法。根据Heisenberg不确定性原理,分别利用长、短时间窗函数获得两组STFT时频表示,最终的时频表示以两组时频表示的乘积确定。仿真跳频信号和遥测振动数据处理结果表明:组合窗函数的STFT时频表示方法能有效提高STFT的时频聚集性,特别适用于跳频信号的分析。
Time resolution and frequency resolution of STFT with fixed length time window function restrain each other,which affects the time-frequency resolution.A method of combining window functions for STFT was proposed to solve this problem.Based on Heisenberg uncertainty principle,by using long,short window function,two sets of STFT time-frequency representation was obtained respectively.The final time-frequency representation was determined by the product of two time-frequency representation.The simulation frequency hopping signal and the vibration signal of telemetry processing results showed that the combination of window functions for STFT improved the time-frequency resolution of STFT effectivelly,which is especially suitable for frequency hopping signal analyse.
出处
《探测与控制学报》
CSCD
北大核心
2010年第3期43-47,共5页
Journal of Detection & Control
基金
国家自然科学基金项目资助(50405023)
