Analytical Methods for the Study of Periodic Phenomena Obscured by Random Fluctuations

The interest shown of recent years in studying random functions has led to a reappraisal of methods for detecting periodicities. Classical methods, such as Fourier analysis and standard statistical methods, work well for truly periodic series; but when a periodicity is obscured by random components, as is often the case in biological phenomena, such techniques have severe limitations. The recently-developed discipline known as "statistical communication theory" has, however, proved most valuable in this subject, and it is from the standpoint of this theory that this article is written.