利用多正弦窗谱分析检测色噪声中的谐波信号

Detection of the Harmonic in Colored Noise Based on the Sinusoidal Multitaper Spectral Analysis Approach

利用最小化代价函数推导了一种谱分析的多窗口方法 ,得到了一组彼此正交的、具有简洁正弦窗形式的多窗口序列。以此为基础 ,结合统计分析的假设检验方法 ,可以利用多正弦窗谱分析技术对色噪声中的谐波信号进行检测 ,准确地估计信号包含的谐波频率及各个谐波的幅度。仿真实验证明了该方法的有效性和正确性。

A multitaper spectral estimator that has an intuitionistic and analytical expression is built based on the minimization of a cost function. The tapers presented in this paper are orthogonal to each other and have the expression of sinusoids. With the approach of hypothesis test, the sinusoidal multitaper spectral estimator can be used to detect the harmonic in colored noise, and to estimate the frequency and magnitude of each harmonic exactly. The validity and the correctness of the approach are verified by the computer emulation experiment.