11/2维谱估计在直升机声信号特征提取中的应用  

Application of One and Half Spectrum Estimation in Feature Extraction of Helicopter Acoustic Signals

作　　者：张浩[1] 高勇[1] 

机构地区：[1]四川大学电子信息学院,成都610065

出　　处：《数据采集与处理》2008年第4期476-480,共5页Journal of Data Acquisition and Processing

基　　金：总装预研基金资助项目

摘　　要：提出一种基于112维谱估计的直升机声信号低频线谱特征提取方法。112维谱是高阶谱的一个特例,它具有几个特殊的性质,这些性质使其能抑制高斯噪声、加强谐波信号基频分量以及剔除非二次相位耦合的谐波分量,因此在处理线谱估计问题时112维谱估计要比传统的功率谱估计方法具有更好的性能;同时与双谱估计等常用的高阶谱估计方法相比,112维谱估计大大简化了计算。通过对实测直升机噪声数据进行仿真,结果证明,将112维谱估计用于直升机声信号的特征提取可行、有效。A method for extracting the low frequency line spectrum of helicopter acoustic signals is presented based on the one and half spectrum estimation. One and half spectrum is a special case of higher-order spectra and has some inherent properties. By these properties, one and half spectrum can restrain the Gaussian noise, enhance the fundamental frequency component of the harmonic signals, and eliminate the non-quadratic phase coupling harmonic components. So when dealing with the line spectrum estimation problems, one and half spectrum estimation has better performance than the conventional methods for the power spectrum estima- tion. Meanwhile, one and half spectrum estimation saves computing time compared with the higher-order spectra estimation methods like bispectrum estimation. Simulation results from actual helicopter noise data show that the method is feasible and efficient.

关 键 词：高阶谱 1 1/2维谱估计 直升机声信号 特征提取 

分 类 号：TN911.6[电子电信—通信与信息系统] V217.1[电子电信—信息与通信工程]