基于空间时频分布的欠定DOA估计

波达方向（DOA）估计是阵列信号处理领域的热点问题，但经典的DOA估计方法通常要求阵元数大于源信号个数，即满足超定条件。而在实际中往往面临的是源信号个数大于阵元数目的欠定条件。提出了一种基于空间时频分布的MUSIC扩展算法，该算法通过将空间时频分布矩阵进行拓展，实现了欠定条件下的DOA估计。相比时频MUSIC算法，该算法能同时适应超定和欠定条件；相比已有的欠定DOA估计方法，该算法在保证DOA估计精度的情况下，放宽对源信号稀疏性的要求，同时降低对快拍数的要求。仿真实验结果表明了本文算法的有效性。

Underdetermined Direction of arrival Estimation based on time-frequency distributions

In the field of array signal processing, direction of arrival estimation has been an popular research when the sensors are larger than source signals, which is so-called overdetermined case. But the signal context which we encounter in practice is always unsatisfied. To solution the problem, this article provides an extension of the MUSIC in time-frequency domain algorithm to the underdetermined case by expanding the dimension of the STFD matrices, giving rise to the time-frequency distribution underdetermined MUSIC (TFDU-MUSIC) algorithm. Compared to the existing time-frequency MUSIC (t-f MUSIC), the proposed TFDU-MUSIC can be applied the underdetermined case. Resorting to the time-frequency distributions, the proposed algorithm has advantages over the existing underdetermined DOA estimation methods including not requiring the source sparseness, resolution of angulation estimation as well as robustness to the number of snapshots. Simulation results confirm the validity and high performance of the proposed algorithm.