用于法拉第旋转测量综合的MUSIC
MUSIC for Faraday Rotation Measure Synthesis
摘要：
法拉第旋转测量（RM）合成需要从受限于有限波长范围的测量中恢复法拉第色散函数（FDF），这是一个病态的反卷积问题。在这里，我们提出了一种新的反卷积方法的基础上的多信号分类（MUSIC）算法的扩展。该方法的复杂度和速度取决于观测极化协方差矩阵的特征分解。我们数值表明，对于高到中等信噪比（S/N）的情况下，RM-MUSIC方法是能够恢复的法拉第深度值的紧密间隔对薄RM组件，即使在FDF的峰值响应的情况下，是两个输入RM组件之间的RM范围之外。这个结果是特别重要的，因为标准的反卷积方法的基础上RM-CLEAN系统地失败，在这种情况下，由于其贪婪的机制，用于提取RM组件。对于低S/N的情况下，RM-MUSIC和RM-CLEAN方法提供了类似的结果。
英文摘要：
Faraday Rotation Measure (RM) synthesis requires the recovery of the Faraday Dispersion Function (FDF) from measurements restricted to limited wavelength ranges, which is an ill-conditioned deconvolution problem. Here, we propose a novel deconvolution method based on an extension of the MUltiple SIgnal Classification (MUSIC) algorithm. The complexity and speed of the method is determined by the eigen-decomposition of the covariance matrix of the observed polarizations. We show numerically that for high to moderate Signal to Noise (S/N) cases the RM-MUSIC method is able to recover the Faraday depth values of closely spaced pairs of thin RM components, even in situations where the peak response of the FDF is outside of the RM range between the two input RM components. This result is particularly important because the standard deconvolution approach based on RM-CLEAN fails systematically in such situations, due to its greedy mechanism used to extract the RM components. For low S/N situations, both the RM-MUSIC and RM-CLEAN methods provide similar results.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
