Phase Dispersion Minimization

Phase dispersion Minimization (abbreviated PDM ) is a data analysis technique, which is determined from a time series measurement of periodic components. It is particularly used when the missing data having periods of non- sinusoidal vibrations, unfavorable timing cover or other disadvantages, which prevent a Fourier analysis. The method has been described primarily by Stellingwerf (1978). PDM often comes in astronomy and physics apply.

Method

Minimization is the phase dispersion of a variant of the folding data. Core of the method is repeated Try or estimating a time period, and then overlaying the data in portions corresponding to the length of the estimation period. The data are folded into a phase plot. Does the estimation period of the true period of the data, the result is in the phase diagram measurements distribution according to a relatively simple function. If this is not the case, then the measured values ​​are distributed randomly.

In order to make this kind of result analysis, PDM divides the phase diagram into several sections and each calculated the variance of the measurements in these subsections. These subsections can optionally overlap each other to get a better coverage of the phase. The variances are summed and set in relation to the variance of the overall measurement. In case of a match the estimation period to the real quotient of the variances is minimal. Not matching periods results in a ratio of approximately one. A graph in which this ratio can be applied over the estimation period, provides the likely positions of periods.

An alternative approach to the division into sub-sections is the consideration of the difference between adjacent measurements. Is the estimation period in accordance with the true, as ordinate values ​​of the measurements are adjacent, and the summation of the differences will be minimal.