Pulsars and neutron stars/Observing pulsars

From Wikibooks, open books for an open world
Jump to navigation Jump to search

Introduction[edit | edit source]

Radio observations[edit | edit source]

Getting observing time for pulsar observations[edit | edit source]

Radio telescopes[edit | edit source]

The basic structure of a typical radio telescope consists of:

  • the Antenna
  • a set of receivers
  • a conversion system
  • digitisers
  • signal processors

Receiver systems[edit | edit source]

A radio telescope receiver captures the signal reflected from the antenna and amplifies the signal.

Coupling noise into the system[edit | edit source]

Backend systems[edit | edit source]

Incoherent dedispersion[edit | edit source]
Coherent dedispersion[edit | edit source]

Carrying out observations[edit | edit source]

Basic observing setup at the Parkes telescope

Types of observations[edit | edit source]

Baseband recording[edit | edit source]

Search mode[edit | edit source]

In search mode the observed sky is averaged over a specified integration time and then recorded to disk normally as a 1, 2, 4 or 8 bit value. The data are usually separated into frequency channels and sometimes in 2 or 4 polarisation states. Various data formats exist for storing search mode data. These include:

In PSRFITS the data are recorded as follows. Each data table contains NSBLK samples of data. Each block contains NPOL polarisation states. Each polarisation state contains NCHAN channels of data.

The Parkes Multibeam Pulsar Survey recorded 96 channels of data with a sampling time of 250us with a 1 bit digitiser. Recent surveys have many more channels and use a faster sampling time (64us is common) and often make use of 2 or more bits.

Fold mode[edit | edit source]

Example of creating a folded pulse profile by summing individual pulses. Image produced by Charlotte Ward.

In fold-mode observations the time series from the telescope is folded at the known period of the pulsar. If the period is correct then each pulse will be added together to produce a folded pulse profile. The animation shows the signal-to-noise ratio increasing as more and more pulses get added together. Folding can be carried out in software on a "search mode" or "baseband" data set, but most major observatories have hardware backend systems that fold the data online and the observer only obtains the resulting fold-mode data file.

Clearly this method only works if the pulse period is known sufficiently accurately. If not, then the pulse will appear to drift in phase over time and the resulting pulse profile will become smeared. To counteract this affect, it is usual to only fold relatively short segments of data in real time (~1 minute in length). These segments are known as subintegrations. These files then get archived. If the folding period is slightly incorrect then offline software can re-align the subintegrations (note that any smearing that occurred during a single subintegration is not recoverable).

Pulsar fold mode files are generally stored as follows:

  • Every subintegration is made up of a large number frequency channels (Nchan)
  • Every frequency channel contains up to 4 polarisation states
  • Every polarisation state contains a pulse profile (which represents the pulse shape defined by a defined number of bins)

Dealing with RFI[edit | edit source]

High-energy observations[edit | edit source]