[time-nuts] GPS week rollover

Bob kb8tq kb8tq at n1k.org
Sat Mar 30 11:30:27 UTC 2019


Hi

If you want it to behave like a signal generator ( variable output level) that also is
another rabbit hole to run down. Since the signals are well below the noise after
full spreading (and not all that strong before spreading) it takes some care to get
that right. On a production basis, testing sensitivity *might* be something you would
want to do…..

There *is* the good old “hook up an antenna and see if it works” approach. In an era 
where not all systems are visible from all parts of the world (or are not yet launched) 
that has some limits ….

Bob

> On Mar 29, 2019, at 9:46 PM, jimlux <jimlux at earthlink.net> wrote:
> 
> On 3/29/19 12:22 PM, John C. Westmoreland, P.E. wrote:
>> To All:
>> I have a related question; hope it's OK:
>> What are these companies using for their "GPS Simulation/Simulators?".
>> Is it actually simulating the RF downlink or is a direct data stream
>> injection into a tap into the receiver??
> 
> RF.
> GPS simulators, recorders, players are a standard thing.
> 
> They range in complexity - some generate the entire sequence "offline" and store it in a big memory, then just play it back. Others can run it in real time and take dynamic inputs (e.g. from a simulated airplane flight).  Some have variable levels.
> 
> The tricky parts are:
> you have to do Doppler too, so it's not just hooking up 1 Mchip/sec PN codes to a mixer and a 1575.42 MHz oscillator.   That said, because many receivers use 1 or 2 bit digitizers, you can make some shortcuts.
> 
> The recorder/playback ones are the simplest - they are literally a 1 bit thresholder running at a suitable rate hooked up to a LNA and the playback plays back the 1 bit through the same filter, and trusts in the harmonics coming through.
> 
> The fancy ones (Spirent is one mfr) are very nice - they're basically a real time waveform generator with a bunch of software, implemented in an FPGA - they can do hundreds of satellites on multiple frequencies, with multipath, relativistic effects, etc.  (Like many software defined widgets, you pay to "unlock" various features).
> 
> 
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.





More information about the Time-nuts_lists.febo.com mailing list