[RC5] Re: Golumb Stuff

Marc Sissom marcus at dfwmm.net
Tue Mar 3 19:55:00 EST 1998

> Thanks for the mini-FAQ... it answered lots of questions. However, I have
> two more:
> a) What makes anyone think that radio dishes should be spaced out in integer
> intervals? That seems very odd. You want a VLA with all inter-dish
> intervals different, wouldn't you want to use all real numbers? Some sort
> of Fibonnacci sequence, at least?

The "units" are "variable". It is the inter-antenna spacing "ratios"
that make the difference. Typically a "unit" is one wavelength of
whatever frequency is of interest. If an antenna array is to be spaced
by 0, 1, 3, etc, These are not feet, meters, or miles, but numbers of
wavelengths or 1/2 waves or 1/4 waves... They might be spaced at 100,
300 miles, or 100, 300cm or 100*1.2345cm, 300*1.2345cm, 1, 3 waves.

Depending on which units you choose, you'll get integers, or

The integer relationship comes from the desire to gather phase infor-
mation from the signal. One cycle, one wavelength, two, three, etc.
It is possible to correct or otherwise account for variations in the
spacing from the ideal, but any such variation results in a reduced
amount of information that can be gathered from the signal.

Effectively the quality of the antenna array is reduced and so there
are resources being wasted. In other words, if a given array of X an-
tennae is not spaced optimally, you could do just as well with an
array that is made of less than X separate antennae. So...to do the
same effective job as a poor quality big array, you could instead
build a smaller(cheaper) ideal one.

Either way, you get the most bang for the buck by building ideal
structures. So we get to do the math! We get to help improve the
quality of research. We get to help everyone get more bang for
their/our research tax dollars! Who'd a thunk it?

Definitely a sexy project.
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest

More information about the rc5 mailing list