Showing posts with label calibration. Show all posts
Showing posts with label calibration. Show all posts

Wednesday, 19 March 2014

North Celestial Pole

In a LOFAR system, a "beamlet" is a pointing direction and a frequency channel (which is called a subband). In some ways, you can think of it as a  "radio pixel" on the sky. These can be static, pointing at a fixed direction, or tracking, where they follow a particular astronomical object as it rises, travels across the sky, and then sets.

This rising and setting, and the motion of the stars generally, is dominated by the rotation of the Earth. Although some objects move about in the sky, due to orbits or proper motion, it is the 24-hour cycle of the Earth that is the dominant effect.

However, there are two points on the sky which do not move very quickly nor very much (although they do still move!). These are the Celestial Poles. These are the points at infinity that are projections of the Earth's rotation axis. Check the Wikipedia article for a full description.

So, in principle, if you were to point a beamlet at the North Celestial Pole (which is the one we see from KAIRA), then you would expect the power level measured to be absolutely flat. There is radio power coming from that direction, but it should remain constant, as it is only rotating within that radio pixel.

It is sort of flat... but not quite!

Here's an example:


The variations come from several things. Tall vertical spikes are radio-frequency interference. And the fuzzy-noisy patches can be from scintillating sources going through the side lobes.

More generally, gradual variations can be from larger areas of Galactic emission and discrete sources passing through the sidelobes of the beam pattern. And then there are subtle variations caused by snow on the antennas, temperature variations, receiver stability, etc.. These are much smaller, but are still present.

As a result, it can be quite a challenge to calibrate data from the instrument.

Such is the fun of radio astronomy!  :-)

Wednesday, 13 November 2013

Subband to frequency checking

On Monday, we reported on the checking of the cells within the HBA tiles. As part of the same test, we also checked on the frequencies. Although there had been some discussion regarding the registration between the subband and the frequency due to some confusion related to the documentation, we had this pretty well resolved a while ago. However, there is nothing like actually checking it for real. Therefore, as part of the general testing day, for the cells, polarisations and other things, we checked of the frequencies as well.

To do this, we built a test aerial. This was a crude dipole made with some wire, a broom and a snow-post. The coax lead from the dipole went back into the site office, where we had a calibrated signal generator.

Our hi-tech aerial! (Photo: D. McKay-Bukowski)

Using this, we could inject signals and look at the corresponding response. Here is one example output, using RCU mode 5:


The colour indicates signal strength. Red is strong, yellow is medium and blue is weak. Of course, most of the spectrum is weak. The injected signals, in time order, were 149.902, 150.000 and 150.097 MHz. Each sample is 10 seconds in duration. We ran each for a about 30 or so seconds, with a similar gap in between.

As you can see, there was the best response at exactly 150.000 MHz, with no spill into adjacent channels. Additionally, the signals injected on the subband boundaries do tend to appear roughly equally in both channels.

Like our other tests, this came as no surprise. However, it is always useful to double check all aspects of the measurements and verify the data integrity.

Tuesday, 12 November 2013

Shifting snow

As mentioned in yesterday's post, we carried out a number of tests on the HBA. Some of these required actually getting into the tile itself (for example, to carry out the polarisation checks). And of course, doing the dipole and RF-clock tests is always easier when the tile is clear from snow. So, although the small amount of snow does not significantly affect our radio measurements, we had to clear it off anyway.

Today's photograph is a shot of just that... yours truly shifting snow off HBA #H25.


Shifting snow. (Photo: P. Martin)

Monday, 11 November 2013

Cell registration

Data Monday...

There has been some discussion recently regarding not just polarisation for the HBA, but also cell registration too. As we were already conducting some other tests, we decided to check on this too. In other words, checking the registration of the individual antenna cells within the tile itself.

The cells within the tile are numbered 1-16, starting in the top left and working across. The position offsets of these are set using the iHBADeltas.conf file.

Although we have been reliably using the system, it has not actually been tested directly. To do this, we created a temporary dipole (some wire, a broom and a snow post) and also a screen (kitchen foil between cardboard, taped to an HBA lids' pallet). Using these, we could both transmit into and block signals for a given cell in the tile.

We also made use of some special rspctl commands to disable all the tile elements, except one. This allowed us to check each cell individually.

The only problem in this was getting the snow off the tile.

A home-made RF block. (Photo: D. McKay-Bukowski)
Of course the test was completely successful, but it is always very gratifying to get such confirmation. The configuration is as follows:


1234
5678
9101112
13141516

Cell #1 is located at the western-most corner of the tile. The cables for the tiles exit underneath cells #14 and #15. Note that this is the not the usual layout for HBA tiles in a LOFAR station. It is just that KAIRA has a slightly different orientation in order to accommodate EISCAT experiments.

Thursday, 18 July 2013

The challenge of calibrating

Calibration can be tricky business. Especially when you get interesting scientific events occurring during the calibration process. Recently, we were attempting to measure the on/off source fluxes to estimate KAIRA's sensitivity. Two problems crept in: ionospheric scintillation and ionospheric opacity.

In the plot, the vertical axis is labelled "beamlet number", but it is linearly-proportional to frequency. The horizontal axis is time. Red is strong radio emission and blue is very weak. The two main blocks in the middle of the plot are Cygnus A and Cassiopeia A.

If you look at the top of them (around beamlets 320-400), you can see a "colourful fringe". This is ionospheric scintillation. We know about this and the solution is to time-average to smear it out.

However at the lower frequencies (roughly beamlets 50-130) there are also some odd effects. In particular, around samples 7000-8000 there is a large drop in radio strength from Cygnus A. This is caused by increasing electron density in the ionosphere (around 80km up).

In this plot it is pretty obvious, but it certainly caused us to puzzle over the rather odd results when we were trying to figure out why the Cygnus A measurements gave such completely different ones to the Cassiopeia A equivalent.

Wednesday, 13 February 2013

Pointing calibration for the 2012 DA14 event

In order to carry out the observations of 2012 DA14, we must carefully check the radar pointing calibration. Remember that this asteroid will fly past quite quickly, so it is essential to check that the alignment is correct.

To do this, we plot the anticipated asteroid trajectory. We then search for radio calibration sources (point-like radio "stars") that are in the sky and work out when they would cross the asteroid trajectory.


The trajectory of the asteroid, as seen from the EISCAT-UHF.site. The circles are the locations on that trajectory where bright radio calibration sources pass. These will be used to carry out the pointing accuracy calibration tests ahead of the actual observations. (Plot by J. Vierinen)

We then can observe these sources as they are in that position. Because these are the anticipated positions of the asteroid, we can check to make sure everything is ready for the critical observations.

Monday, 23 July 2012

Antenna cross correlations

We are now routinely collecting data from KAIRA's High-Band Antenna (HBA) array. This operates in one of several receiver unit (RCU) modes which, with the use of filters and clock rates, allow us to get at different Nyquist zones in the spectrum. Most of the initial experiments that we are doing at with RCU mode 5 (which is approximately 110-190 MHz). There is one RCU per signal path. That is, one RCU per polarisation per antenna. For the KAIRA HBA array, we have 48 antennas (the tiles) and each one has 2 polarisations (linear X and Y) and we have 96 RCUs to process all of these.

When data is collected, it is split into frequency channels called subbands. Additionally, it is possible to run a small correlator in the RF-container, which will determine the cross-correlations for each pair of

The following plot shows an antenna cross-correlation (ACC) matrix. The two axes are the RCU numbers, from 0 to 95 (from left to right and from top to bottom). The shading is the level of correlation. The auto-correlations have been set to zero to make the plot more useful. Auto-correlations are what you get if you correlate an RCU with itself. Because this obviously will have a very strong correlation, these data would otherwise dominate the dynamic range of the plot, which is why they are clipped to zero for normal viewing.


KAIRA HBA(5) subband 240, just after midday UTC on 19-Jul-2012.

In the grand scheme of things, these data are really just an intermediate step along the way to getting the system calibrated and producing data streams from on-sky beams, or all-sky images. However, it is an important matter and being able to produce this sort of display is a very encouraging step towards the realisation of a fully working system. It represents the culmination of a lot of hard work and effort by a lot of people, from SGO, ASTRON and elsewhere and everyone involved should take heart that we are getting close now and that their endeavours are appreciated.

Calibration and commissioning work will be continuing this week. Hopefully we should have some even more interesting results to show soon!

Sunday, 22 July 2012

Calibration life

While the building and construction on the LBA continue outside and the installation and setup goes on within the RF-container, there is another side to the work at KAIRA. Within the Lehtinen Barracks, some of the team are working on calibration and software.Despite what the crew outside might tell you, this is actually pretty intense work as we wrestle with all sorts of problems, software packages, operating systems, configurations and, of course... trying to make sense of the data.

The Lehtinen Barracks is a convert railway carriage, fitted out with desks and chairs and work benches. Scattered about are computers, cables, power supplies and various notes and drawings. The heater is running, but still the cool Arctic air gets in. And, when the wind blows, the entire barracks rocks gently back and forth, as it is suspended above the ground. The place smells of a strange mix between coffee and deet (= insect repellent). There are cups (both paper cups and traditional Finnish timber ones) strewn about and scattered here and there are (usually empty) packets of Angry Birds sweets.


What you don't get from looking at this photograph is the sound... do you remember the soundtrack from the original version of the computer game Die Siedler / The Settlers (1993 Amiga version, in-game music by Haiko Ruttmann and Markus Kludzuweit — LINK). Well, this is set to continuous loop and has been playing for several days now non-stop. With the sound of the wind, as well as digging, hammers, sawing and various other noise from outside, it is somehow vaguely appropriate. And, given that we have barracks, lots of stone and plenty of wood, the catch phrase whenever we need something is to add "... and we also need a granary!"