ISSUE 382|COMPUTER SHOPPER|DECEMBER 2019 109
known to have potentially habitable
planets–makes useof the fact that
the AllenTelescope Arraycontains 42
separatedishes six metres in diameter,
insteadof one huge one.This allows
the signals from multiple antennasto
be phase-shifted and addedtogether in
order to generateavery narrow fieldof
view,just 30 arc seconds wide,and to
steer it under software control,to the
area of interest. This isahugely
processor-intensive task.
Then we cometo the breaking up
of abroad radio signal in orderto look
forvery much narrower signals. This is
carried outby aprocess calledaFast
Fourier Transformation and, bearing
in mind that the three gigahertzof
bandwidth hasto be split into72
million segments, it’s perhaps not
toosurprising that thistoowill give
ordinary computing hardwarearun
forits money.
And finally,there’s the needto
scrutinise those signalsto see if any
have potentially been generatedby
extraterrestrial civilisations. This
involves discounting any that could
have been generated on Earth–itself a
huge multistage process–while
keeping any that have the expected
characteristicsof an artificially
generated signal. When we also
remember thatasignal generated on
another astronomical body might
exhibit achange in frequency with
time dueto avariable Doppler Shift, it’s
clear that suchasignal might migrate
from one channelto another,to
another.Again, we can see somethingof the enormityof the requirementto
conduct all this in real time.
Youmight reasonably expect that
this is ajob forasupercomputer,but
actually standard PCs are used.
Where theydiffer from the one on
your desk is that these PCs have
dedicated hardwareto offload the
heavy number crunching. Included hereare GPUs, perhaps nottoosurprising
since these have recently become
almost standard components in the
world of high-performance computing.
Perhaps less well known is the useof
FPGAs. Otherwise known as Field
Programmable GateArrays, these
devices use softwareto reconfigure
the hardwareto perform specificFROM SETI TO METI
Active SETI, otherwise known as Messagingto Extra-Terrestrial
Intelligence (METI), isasomewhat controversial alternativeto
the SETI approach discussed in this article.Insteadof lookingfor
signals from extraterrestrial intelligence,its aim isto announce
our presenceto any civilisations that might be interested.
While opinions on the wisdomof this approach differ in the
SETI community–with some arguing that agreement should be
obtained from the worldwide scientific, political and humanitarian
society before any such attempt is made–METI initiatives have
already taken place.
The first and most publicisedof these was the additionof a
plaque to the Pioneer 10 and 11 spacecraft that were launched in
1972 and 1973, and have subsequently become the first man-made
objectsto leave the confinesof the solar systemforinterstellar
space.The plaquesof gold anodised aluminium are engraved with
line drawingsof humans witharather crude drawing indicating
the locationof our home planet.
While the Pioneer plaques are surely very unlikelyto be
encounteredby extraterrestrials, several radio signals have also
been sent, starting in 1974 withamessage sent from the huge
Arecibo radiotelescope in PuertoRico in the directionof the
M13 star cluster.
It was alow-resolution monochrome digitised image containing
astylised pictureof ahuman, plus several other elements which, it
was argued, would be meaningfulto intelligent civilisations.
Given the date, the similarityof that imageto games on early
home PCs isn’ttoosurprising. That image will reach its destination
in around 25,000 years. If that time period comes as somethingof a
relief, you’ll be less pleasedto hear that one METI signal, sent in
1983, will already have reached its target, and several others will do
so in the nextfewdecades.⬆The SETI Institute’sAllen Telescope Array
in California currently has 42 six-metre
dishes, but it’s plannedforthis eventually
to be increasedto 350Photo: Seth Shostak/SETI Institute
⬆This low-resolution
image –the actual message
was monochrome–was
transmitted in 1974towards
the M13 star cluster