A Study of the Spectral Analysis of the Unequally Spaced VLBI Observational Material on the Basis of Lomb Algorithm

In this article, the problem of the appearance of false signals in the spectral analysis of unequally spaced sampling is investigated. After quasi-multi-periodically sampled simulative material and VLBI observational material are analyzed by Lomb's method of unequally spaced spectral analysis, we draw the conclusion that quasi-multi-periodic sampling may yield false spectral peaks. An empirical formula of the distribution of positions of false spectral peaks is derived, and it is then verified statistically. Concrete application of the formula is discussed. While the conclusion is based on VLBI observational data, it can also be applied to any quasi-multi-periodic sampling material.