[time-nuts] Need advice for multilateration setup

Magnus Danielson magnus at rubidium.dyndns.org
Wed Apr 8 15:21:48 EDT 2015

Hi Jim,

On 04/08/2015 12:46 AM, Jim Lux wrote:
> On 4/7/15 11:33 AM, Magnus Danielson wrote:
>> Hi,
>> O
>> One might look at the available frequencies and see if there is a
>> telemetry band available which allows wider bandwidth. For the
>> application, I don't see that very much transmitted power is needed.
> If the OP is a licensed amateur radio person, then choosing one of the
> low microwave ham bands would be easy.  Parts to generate a carrier and
> BPSK at 2.39-2.45,3.3-3.5, 5.6-5.8 GHz are cheap and readily available.
> You might be able to get away with a VCO and no crystal as the
> transmitter, but even if you can't, there's tons of PLLs out there that
> will nicely lock to a crystal and are cheap.
> You might want to do a link budget and see how much power you need to
> radiate, so that you get a decent SNR at the receiver.
> free space path loss between isotropic antennas (in dB)
> = 34  + 20 log10(freq in MHz) + 20 log10(distance in km).
> 1km at 3 GHz is 34+69 = 103 dB.
> If you radiate 1 mW (0dBm) from an omni (a piece of wire), you'll see
> -103 dBm at the input to your receiver, which is a fairly healthy
> signal.  A detection bandwidth of 10 Hz would have a noise floor of -164
> dBm before taking into account the receiver noise, but even if the
> receiver is terrible, you're still looking at tens of dB SNR with a very
> simple transmitter.

Indeed. I realized that without doing the numbers, so I think the focus 
could be in how to realize a simple and light transmitter. A small FPGA 
will suffice for the code-generation. It will be essentially empty.
Re-cycling the GPS C/A codes should be trivial. It should not be too 
hard to build the receiver side too. It's essentially the same as 
building a GPS receiver.

>>> That's a very good argument for higher chiping rates.
>> I expect that the launch is a bit challenging for the tracking loop.
> If you're trying to track in real time, certainly.  If you're doing post
> processing, less so.

Fair enough. If you know you can track it in real time, then you know 
you can do it in post-processing.


More information about the time-nuts mailing list