基于广义解调平滑能量分离算法的瞬时频率估计

发布时间：2018-03-27 12:44

本文选题：信号分析　切入点：瞬时频率　出处：《振动工程学报》2014年02期

【摘要】：平滑能量分离算法能够跟踪调幅调频信号的瞬时频率,结合广义解调和复解析Gabor滤波器的优点能够克服平滑能量分离算法只适用于单分量窄带信号以及对噪声敏感的局限性。理论分析了负频干扰对传统广义解调多分量分离方法的影响,在此基础上提出了一种新的基于广义解调的平滑能量分离算法,该方法利用广义解调将非平稳信号转化为准平稳信号,再通过复解析Gabor滤波器对其进行滤波以达到单分量信号分离的效果,分离出来的准平稳信号无需进行逆广义解调,直接采用平滑能量分离算法求取瞬时频率,经过频率补偿得到原始信号的瞬时频率。仿真和试验结果表明该方法能够克服负频率干扰,且比传统方法具有更高的解调精度,进一步扩大了平滑能量分离算法的应用范围。
[Abstract]:The smoothing energy separation algorithm can track the instantaneous frequency of AM signal. The advantages of generalized demodulation and complex analytic Gabor filter can overcome the limitation that the smoothing energy separation algorithm is only suitable for single component narrowband signals and is sensitive to noise. This paper theoretically analyzes the negative frequency interference to the traditional generalized demodulation multi-division. The effect of quantitative separation methods, On this basis, a new smooth energy separation algorithm based on generalized demodulation is proposed, in which the non-stationary signal is transformed into quasi-stationary signal by generalized demodulation. Then it is filtered by complex analytic Gabor filter to achieve the effect of single component signal separation. The separated quasi-stationary signal does not need to be demodulated by inverse generalized demodulation, and the instantaneous frequency is obtained directly by using the smoothing energy separation algorithm. The instantaneous frequency of the original signal is obtained by frequency compensation. The simulation and experimental results show that the proposed method can overcome the negative frequency interference and has a higher demodulation accuracy than the traditional method, which further expands the application range of the smoothing energy separation algorithm.
【作者单位】：华南理工大学机械与汽车工程学院;
【基金】：国家自然科学基金资助项目(51005078;51075150) 广东省博士启动基金资助项目(S2011040003851) 华南理工大学中央高校基本科研业务费资助项目(2012ZM0045,2012ZM0044)
【分类号】：TN911.7

【参考文献】