高阶时频变换理论与应用

High-order Time-frequency Transform Theory and its Applications

非线性信号处理与特征提取是时频信号处理领域中的热点问题。传统时频分析方法受限于海森伯格不确定原则,在分析该类信号时存在较为严重的能量发散问题,导致分辨率不佳,无法为非线性信号处理与特征提取提供精确的时频信息。为改善传统时频分析方法的能量聚集性,提升其时频分辨率,提出了一种高阶同步抽取变换算法,首先计算多个短时傅立叶变换,之后用以估计高阶瞬时频率算子,并结合短时傅立叶变换谱提取信号的非线性瞬时特征,从而能够为信号中的非线性特征提供高分辨率的时频谱。仿真分析验证了提出方法在时频特征刻画、能量聚集性改善以及抑噪方面的性能。实验分析验证了提出方法能够用于复杂多分量非线性信号处理,对于信号处理理论拓展具有重要意义。

Non-stationary signal processing is always a hot topic in the signal processing fields.Traditional time-frequency analysis methods are limited by the Heisenberg uncertainty principle,which cannot provide accurate time information for highly time-varying signals.In order to improve the energy concentration of the time-frequency analysis method,a high-order synchroextracting transform is proposed.By estimating the instantaneous frequency of the signal,the proposed method can characterize the non-stationary characteristics of complex signals.Simulation analysis and experimental analysis verify the performance of the proposed method in time-frequency feature characterization and noise suppression.