Sound Familiar?


We sometimes analyze datasets gathered elsewhere. Here is a dataset gathered during the setup for a big public event at Jodrell Bank i.e. during a time when the radio environment was deliberately not well controlled (click on plots to zoom on this page):


The big spikes in the bottom (de-dispersed) plot correspond to strong signals from the Crab pulsar, the object of scientific study in this case. The top time-series plot of 60 secs duration shows some interesting features:
  • There are some small positive spikes that appear regularly at 1 sec intervals. It turns out these were deliberately injected into the signal chain in this case by the scientists/engineers in order to provide a timing reference.
  • In addition, and more noticeable, are some very big drop-outs in the signal, starting at around 16 sec and 25 sec. Zooming right in on some of these in the top plot below shows that they are sequences of fairly short duration:

At first glance, it was not very clear what these were. However converting the time series data into an audio signal reveals a familiar sound (at around 16 secs and 25 secs)! Recognize what that is? That is why we go to great lengths to keep the radio environment as clean as possible near our telescopes.

In this case, the system had some sort of active suppression of radio frequency interference signals in place which attenuated the normal signals too and resulted in a kind of "negative imprint" of the interfering signal (analogous to the cavity left behind after certain fossils dissolve, leaving the shape of what was there previously).

Making a single pulse plot of the signal for various dispersion measures (follow link to explain this), we obtain:


The "mysterious" interfering signal shows up strongly in red at DM=0 (and spreading vertically downwards into the image). The science signal we are after from the Crab pulsar appears as the weaker sporadic signals at around DM=57. Zooming in on these (the red bits of the image below):


If we were were interested in looking for strong single pulses (such as Fast Radio Bursts) from more distant locations, we can analyse the same dataset further over a wider range of dispersion measure (DM) "space" to produce the plot below (out to DM=1500):


In this case, no other obvious signals are visible in this direction of the sky and for these 60 secs (but we will keep looking!).



Ć
Jodrell_Crab_time_series.wav
(458k)
Jasper Horrell,
Nov 21, 2013, 12:25 AM
Comments