For fun (I really can not take this stuff too seriously), I downloaded some ligo data.
And made a program to display and filter it.
This is the raw data coming from the LIGO:
click for large version
These is really nothing to see.
The data is the photon pressure on the mirror in the centre, as I understand from the description.
[quote "Ligo tutorial"]
(see link)
The data are dominated by low frequency noise; there is no way to see a signal here, without some signal processing.
There are very low frequency oscillations that are putting the mean of the L1 strain at -2.0e-18 at the time around this event, so it appears offset from the H1 strain. These low frequency oscillations are essentially ignored in LIGO data analysis (see bandpassing, below).
....
The bandpassing cuts off frequency components below around 20 Hz and above 300 Hz.
[/quote]
So I still got some work to do on the filters, the simple filters I use now do not seem to work.
I don't trust the fourier filter in this case, it can add weird transformation in non-linear data.
I'll come back with more data later. Also with higher frequency data.
But by just looking at the raw data, everything looks like a chirp, or something.
It does not give me the impression that this is a very clear signal.
And it is very suspect that a signal is "found" near the common electric grid frequency.
Looking at the signal itself:
I hope to produce a better, more detailed picture after I got the filters working.
If it is really possible.. ha ha
The black line is a complex waveform, which tries to match both signals.
This signal is filtered, shifted in time, and one is inverted.
If we would filter even more frequencies, the signal would be resembling a perfect sinus
(for both red and green). So the filtering makes these signals more similar than
they actually are.
We can see that only significant part of the signal is only a very small part, in
which both the red and green spike a bit.
Notice that both the red and green signal are most of the time NOT following the black path.
The black path is just a crude average. Only a "microsecond" at 0.01, they are on the same path,
for only one period of the signal. Green and red have different amplitudes at 0.00 to 0.01.
So we have to explain very different amplitudes, if we accept this as a real GW signal.
If we look at the rest of the signal, we can see that the red and green follow a different path.
After 0.01 the signal goes on a bit in the same frequency, as if there is not really an end.
This signal does not stop at 0.02. It is the same frequency as around -0.08.
And this is not even the RAW signal, in which this is completely invisible.
At -0.04 the signals of red and green are clearly different, and before that the signals are
still periodic, but completely separate. In this example I can not see the phase-shift so clearly,
at -0.04. But in this picture there is a clear inversion of red and green at -0.06.
So if we look at only the significant part of the signal, compared to the black, we can see that this is
very small and with a lot of noise. And that after we already removed most of the noise
with our filters. This should not happen.
If we remove the black line, we see two different signals that "spike" with different amplitudes,
at about the same time with a frequency of +-0.008 sec (=125Hz).
So I would rather talk about a "spike" instead of a "chirp".
The noise seems a signal of around +-0.01 sec (100Hz) and +-0.005 sec (200Hz).
If both interfere they can easily produce the spike with that frequency.
I wonder if the LIGO scientists actually measured the resonance frequencies in their system.
From
the LIGO documentation I can read that their lasers follow a zig-zag path of
around 1120 km, where it goes through the laser again (power recycling).
With c/1120= 268 Hz, we can see that this laser recycling can be causing a continuous noise
signal within the range of what we are measuring. It is close to the +-200Hz already.
If we double the time, we get the frequency of the "spike" signal.
Maybe it was caused by the lasers or mirrors warming up.
(The LIGO was still in a warming up phase).
I hope to see more in the RAW signal.
Even if we assume that the black line is the real signal, it is still different from a real GW signal.
The theorized GW signal has a much longer beginning. If we trace it back we should see a periodic
signal going for seconds.
So, even in the beginning of this RAW data investigation,
the sigma of the GW seems very very overrated.
After some time I will post an update on the filter and maybe new conclusions.