A Proposal for a Low Budget, Highly Effective Intergalactic Communicator


By Neil J Boucher, Radio Communications Engineer

July 2000




A proposal is presented here for a low frequency, low cost and yet effective intergalactic broadcast transmitter.


Big gains (in link budget), can be made by simply sending the information slowly. So it is suggested that SETI on a budget would have a very slow data rate (whatever the carrier frequency).




To understand how this proposal works, we will need to examine a few basic radio concepts and to understand the difference between low frequency and high frequency communications.  The essential difference between transmissions in the GHz range and those in the VLF and lower is that to capture a good signal at the high end of the band you need an elaborate antenna (like a huge dish or a complex array).  At lower frequency you can capture the same energy with a simple dipole, consisting of nothing more than a length of wire.


The simplest effective dipole antenna is made of two lengths of wire each a quarter of a wavelength long, with a transmitter or receiver midway between the wires.


Radio Path Loss


Radio engineers define the path loss of a radio link to be the loss over the path in question that would be measured between two dipole antennas.  Lets look at the path loss that would be experienced if we were to attempt to transmit across the “average “ galaxy of 10,000* light years diameter.


This can be calculated from the formula


Path loss (in dB) = 20 Log (F)+20 Log (D) +32.5


Where F= frequency in MHz

            D = distance in Km


From the above it is clear that the path loss increases as the logarithm of the frequency, and this should immediately suggest that low frequencies might have an advantage.


So lets list some actual path losses across 10,000 light years.


Frequency (in MHz unless stated otherwise)         Path Loss (dB)


1500        1500                                                                     435.5

150            150                                                                       415.5

15                15                                                                         395.5

1                    1                                                                            372.

10,000 Hz                                                        312.

100 Hz                                                             272

10 Hz                                                               252

5 Hz                                                                 246

1 Hz                                                                 232

0.1 Hz                                                              212


Next we can consider that while high gain antennas are possible at higher frequencies they are not at the “audio” levels being considered here.  If we assume that a large 64 meter dish is being used for the high frequency reception and a simple dipole is used for frequencies of HF (30 MHz) and below, we will have a gain at 1500 MHz of 59 dB, and 39 dB at 150 MHz.  We can allow a gain of 0 at 1 MHz and –10 dB at the lower frequencies (that is the dipoles will be shorter than a quarter wavelength).


Now lets define the source to be a “simple” 10 Megawatt transmitter, fed into a dipole (in order to get a nearly omni-directional pattern…to obtain a true omni-directional pattern a simple array could be used).


We can define a range gain for the proposed links.  The range gain is the net path link budget, including all antenna gains, TX power and RX sensitivity.  The RX sensitivity is not something that is fixed, but can be varied by simply varying the bandwidth (the narrower the bandwidth, the higher the sensitivity).  Let us arbitrarily set the RX sensitivity at –150 dBm.  For the frequencies above we now have link budgets as below


Frequency (in MHz unless stated otherwise)   Path Loss (dB)      Link Budget(dB)


     1500                                                            435.5                       310

150                                                              415.5                       290

15                                                               395.5                        251

1                                                                 372.                            

10,000 Hz                                                   312.                          241

100 Hz                                                        272                             

10 Hz                                                          252                            

5 Hz                                                            246                            

1 Hz                                                            232                            

0.1 Hz                                                         212                            


It can now be seen from the above that only the really low frequencies have a link budget which, exceeds the path loss.  The suggestion therefore is that to transmit across the galaxy (or the universe for that matter) the budget way to go is at low frequencies.


There are other advantages of low frequencies.


First, running a suitable program a desktop PC could easily look at all possible low frequencies for a signal, therefore eliminating the need to guess the magic frequency. 


Secondly the low frequencies would necessarily (almost) be omni-directional and so a broadcast to the galaxy would be sure to be heard by all potential listeners.


Thirdly the transmitters and receivers are simple to construct (although there may be a few challenges with the transmitting antennas).


*Note the Milky Way is about 100,000 light years across so you need an extra path budget of 20 dB if you are to get across it.


Capture Area of an Antenna


If you operate at high frequencies the way to increase the capture area of your antenna is to use a bigger dish or a bigger array.  Either way it is expensive.  At low frequencies all you need is a simple dipole antenna (basically a length of wire), which has a capture area that is proportional to the square of the wavelength.


So if you consider transmitting on 100 Hz and then calculate the path budget, it will be 100 times (or 20 dB better) at 10 Hz, and in turn another 20 dB better at 1 Hz (for a total of 40 dB).


Small Antennas


At very low frequencies a dipole (half wave) antenna can get rather large.  For example at 1 Hz it is 150,000 km long.  However there is an effect that comes to the rescue.  The aperture of a short dipole (that is one very much shorter than a half wavelength) is 0.119 times the aperture of a half wavelength antenna; so such an antenna is less than 10 dB down on a full size quarter wave antenna.  This “gain” does however come at a cost that the radiation resistance of a short dipole is very low and ohmic (resistive) losses in the antenna will be high.   With high ohmic losses the efficiency of the antenna will be low.


Here again we see the advantage of low frequencies.  Superconductors offer zero-resistance only to DC.  Some resistance is experienced by any alternating current, and that resistance increases with frequency.  For different reasons the same is true of very cold low resistance conventional conductors.


No need to Rush


In a recent article in the Scientific American (July 2000) George Swenson of the University of Illinois rightly points out that the link budget is a function of the bandwidth used (because if you tighten the bandwidth you “expose yourself” to less noise).  However he then goes on to suggest a break-neck data rate of 5 bits per second!!!


Remember that if ET sets up the transmitter even to get across the Milky Way, it must be anticipated that it might take 100,000 years for the most distant receiver to get the message; so what does it matter if the message is sent slowly. Instead of 5 bits per second why not 5 bits per 100 googlesplonge (that’s 0.005 bits per second to you earthlings!)? By not being in a rush we have instantly improved the signal to noise ratio by 30 dB!!!!


Incidentally, Swenson’s article provides a very good insight into just how difficult it would be to transmit and receive (with any degree of confidence) a GHz signal across the galaxy.  He concludes that it is not within the technical capabilities of today, and may not be for some generations to come.


What to Transmit


In order to transmit data over low frequencies, we need to be patient as only low data rates are possible.   The upside of this is that the slower the data is transmitted the better will be the link budget as greater RX sensitivity follows. 


Since the message is potentially reaching 100 billion star systems, it is of no real consequence that it is slow.   Lets assume ET decided to use a frequency of 10 cycles/googlesplonge  (which happens to be about equivalent to 1 Hz, because of course, a googlesplonge is about 10 seconds).  A googlesplonge is the agreed intergalactic unit of time.   In order to make the signal conspicuous it could be a binary signal, which has a mark space ratio of 100 googlesplonge OFF and (100 x Pi) googlesplonge ON.


There could be two sets of information transmitted.  The first set could be the basic decoding instructions so that the message could be read.  This set of information could be continuously repeated (lets say the sequence took 500,000 googlesplonge to transmit, then we would have about two months of decoding instruction); this could then be followed by a second set of information that transmitted the galactic stock market rates (or whatever else is considered news) for another 1 million googlesplonge, before reverting back to the training sequence.


Building a receiver


A ground- based receiver would be relatively easy to build as it could use an open wire transmission system such as a power cable or telephone line running a long distance in a remote area.  Some railways may still have their old cables in situ but not working and these would be ideal.


By far the best way would be to use an orbital satellite with a long trailing antenna.  Provided the spacecraft was far enough away from the sun, it could provide a temperature such that the antenna was either super-conducting or so cold that its intrinsic resistance was negligible. 


Building a transmitter


Here again ground-based possibilities exist but again a super-conducting orbital platform would be better.  In particular if the orbit were such that the transmitter was say 1 light year from the parent star, not only would super-conduction of the antenna be assured, but, the orbit itself could be used to calculate the location of the home planet.




Looking at the problem of communicating across the galaxy as an engineer, rather than as an astronomer, it does seem that it is not only practical, but, it can be done with today’s technology and on a budget that would not be excessive.  Both the receivers and transmitters are rather low-tech.  The antennas might prove a bit tricky, but the technology to build them exists.


It would seem that the low budget way to get signal across the galaxy is to use low frequencies and low data rates.  The lower the frequency and the slower the data rate the cheaper it is to do.