next up previous contents
Next: Acknowledgement Up: On-The-Fly observations Previous: Time/Sensitivity estimation   Contents


Comparison

Contrary to tracked observations, the position switched observing mode can be more efficient than the frequency switched observing mode. Indeed, in frequency switch, the same time is spent in the on and off spectrum. When subtracting them, the off brings as much noise as the on. In position switch, the same off can be shared between many ons, in which case the optimal integration time on the off is much larger than on each independent on spectrum. Hence, the noise brought by the off spectrum can be much smaller than the noise brought by the on spectrum.

For frequency switched observations,

\begin{displaymath}
\ensuremath{\sigma_\ensuremath{\mathrm{fsw}}} = \frac{\sqrt...
...th{\mathrm{tel}}}\,\ensuremath{t_\ensuremath{\mathrm{tel}}}}},
\end{displaymath} (46)

while for position switched observations,
\begin{displaymath}
\ensuremath{\sigma_\ensuremath{\mathrm{psw}}} =
\frac{\ens...
...th{\mathrm{tel}}}\,\ensuremath{t_\ensuremath{\mathrm{tel}}}}}.
\end{displaymath} (47)

We thus have
\begin{displaymath}
\frac{\ensuremath{\sigma_\ensuremath{\mathrm{psw}}}}{\ensur...
...map}}}}{\ensuremath{n_\ensuremath{\mathrm{beam}}}}} \right) }.
\end{displaymath} (48)

Position switched OTF is more efficient than frequency switched OTF for
\begin{displaymath}
\frac{\ensuremath{n_\ensuremath{\mathrm{beam}}}}{\ensuremat...
...\ensuremath{\mathrm{on/off}}}\ge \frac{1}{3-2\sqrt{2}} \sim 6.
\end{displaymath} (49)

Moreover, $\ensuremath{\sigma_\ensuremath{\mathrm{psw}}}/\ensuremath{\sigma_\ensuremath{\mathrm{fsw}}} \simeq 0.84$ for $\ensuremath{n_\ensuremath{\mathrm{on/off}}}= 30$, and $\simeq 0.78$ for $\ensuremath{n_\ensuremath{\mathrm{on/off}}}= 100$. Using eqs. [*] and [*], we see that the limit on the maximum number of on per off is set by
\begin{displaymath}
\ensuremath{n_\ensuremath{\mathrm{on/off}}}= \frac{\ensurem...
...math{v_\ensuremath{\mathrm{area}}^\ensuremath{\mathrm{max}}}},
\end{displaymath} (50)

i.e. the ratio of the maximum system stability time by the minimum time required to map a telescope beam.

As for tracked observations, there are other considerations to be taken into account. For extra-galactic observations, the lines are large which excludes the use of frequency switched observations. For Galactic observations, the intrinsic sensitivity of the receivers may make it difficult to find a closeby position devoid of signal. We can still use the position switched OTF observing mode. But we then have to observe the off position in frequency switched track observing mode long enough to be able to add back the off astronomical signal.


next up previous contents
Next: Acknowledgement Up: On-The-Fly observations Previous: Time/Sensitivity estimation   Contents
Gildas manager 2014-07-01