Pulsar Dispersion Measure Dedispersion
Delay $\Delta t = \mathrm{DM} / (2.41 \times 10^{-4})(1/f_{\mathrm{MHz}}^2 - 1/f_{\mathrm{ref}}^2)$ ms; the dynamic spectrum shows the $f^{-2}$ sweep, the dedispersed time series peaks sharply at the true DM and flattens out everywhere else.
WHAT TO TRY
- Tune the trial DM toward the true value: the dynamic-spectrum sweep straightens and the dedispersed series collapses from a smear into a sharp pulse. The peak maximizes at the right DM.
- Watch the f^-2 sweep: low frequencies arrive later because the ionized medium delays them more. That frequency-dependent delay encodes the column of free electrons.
- Overshoot the DM: the pulse over-corrects and smears the other way. The sharpness of the dedispersed peak is the matched-filter estimate of the true DM.