The Data Communications (particularly of HDTV Video) required from the Moon is considerable. It is estimated at 1 GigaByte total information (after error correction). This is 3 to 5 times the old “Encyclopedia Britannia” standard for a “Large” data file (which is no longer considered large at all!). We forget that data communication over a radio link is actually rather expensive and that the 240,000 miles to the Moon makes this cost far greater than it would be for 1000 to 22,000 mile, normal satellite communication.

Modeling the necessary communications channel is an interesting challenge. It is easier to model the link for a satellite in space at comparable distance, for example at the EM L4 or L5 orbital points. In this case, the Galactic background noise is very small (2.7 Kelvin) and a very week signal can be picked up by a good receiving antenna on Earth. But the Moon has much larger radio noise emission! A very directional antenna will seen an average of 213 K noise temperature when aimed at the Moon and needs almost 100 times the transmitter power required for communication from L5. A smaller, less directional antenna will mix the noise from the Moon with a large amount of “Relative Silence” from Galactic space, and have a lower “Noise Temperature”. An antenna with a 2.5 degree beam width, for example, will collect its signal from 25 times the Moon's area, and have a noise temperature of 10 K. But this antenna will also collect only 1/25 of the energy from a transmitter compared to a large antenna with a beamwidth equal to the Moon's visual size.

An much larger antenna, with a beamwidth much smaller than the Moon's diameter, will see roughly constant noise (213K) but collect much more signal. But between the antenna which picks up enough noise from the Moon to match its preamplifier noise level (reasonably 10 - 20 K) and the one “Fully Illuminated” by the noisy Moon, there is NO Signal to Noise improvement with increasing antenna size!

(Fortunately the Moon's “Radio Noise Temperature” is not nearly as high as the Moon's surface temperature, and shows much reduced variation with Lunar day and night.)

A “Phased Array” of antennas can be used to simulate a physically larger antenna – including one of impractical 100 meter to 1km diameter! But the simulation is nearly perfect only when the elements which make up the array are almost touching. An “Oversize Array”, with large spacing between its antenna elements, has considerably different characteristics. The total energy received from a compact source depends only on the total antenna area, and is independent of the spacing of phased components. But the “resolution” for this case can be much greater than for a single antenna of equal area, and even exceed the resolution of a single antenna of equal maximum edge to edge size. The tradeoff in this case is that the “Oversize Array” always contains “Sidelobes” of response which the single large antenna does not need to have. These sidelobes are constrained to the beam width of the individual antenna elements and can each be as small in beam width as the combined main beam of the array.

TWO IMPORTANT POINTS: Two important points are buried in this esoteric discussion: First, the SETI Institute ATA receiver array (offered by this “Partner” of the Google Lunar X PRIXE) is exactly this sort of “Oversize Array” - intended to allow extremely high discrete source resolution - and Second, such an array can have a LOWER NOISE Temperature that a single antenna producing equal signal! (Since most of the sidelobes will not see the Moon.) The noise temperature of this system will probably be at least TEN TIMES smaller than a single antenna with equal signal capture, and require only 1/10 the transmitter power!