多路延迟结构的修正MUSIC算法频率估计

Frequency estimation of modified MUSIC algorithm based on multi-path delay structure

提出将多路延迟数字测频结构与高分辨率谱估计的MUSIC算法相结合,有效地实现了信号频率估计。针对MUSIC算法在低信噪比条件下频率估计性能下降的缺点,研究的修正MUSIC算法通过对接收数据共轭重构,产生新的协方差矩阵,有助于改善特征值的分布,使其具有平均的意义,从而提高了信号频率估计的性能。实验结果证明,相对于常规MUSIC算法,该方法在相同的信噪比和快拍数条件下,具有更高的频率估计精度和稳健性,而且计算量也无明显增加。

A new frequency estimation method based on the digital frequency measurement structure of multi-path delay and high resolution spectrum evaluation MUSIC algorithm is proposed, which can estimate signal frequency effectively. But the frequency estimation performance in low SNR will decrease. This paper gives a modified MUSIC （MMUSIC） algorithm, which can get a new covariance matrix by reconstructing the received data and their conjugatings. This method can change the eigenvalue distributions of the covariance matrix and make it have averaged significance, thus it improves the performances of the signal frequency estimation. With the same SNR and snapshot number, the performances of the MMUSIC algorithm has better frequency estima- tion accuracy than the traditional MUSIC algorithm. Simulation results show that it has better feasibility and robustness, as well as the computational complexity will not increase obviously.