Hello all, I'm a recent recruit to this list, it looks very interesting and I hope I can make a contribution. Though now retired I'm an electronics & radio engineer and until last March worked for u-blox (though not on the GNSS side of the business).

Looking at their product integration guides they don't mention antenna match and cable length. They do suggest inserting a SAW filter between the antenna and module especially with an active antenna, and of course the group delay of a SAW filter makes it equivalent to quite a long length of cable - at least 30 cm a nanosecond. Multiple reflections will be attenuated at each reflection - if the return loss is say 6 dB then you lose that much at each reflection, plus the attenuation in the cable itself. So the effect of the cable plus mismatches will be to add an extra channel impulse response that will be quite short as the various echoes decay. Given that the system is working with wanted signals well below thermal anyway, a couple of reflections plus a dB or two of cable loss would make the largest echo ~15 dB below the wanted signal. I suspect that this will cause rather small error in the device's estimate of pseudo-range but this will be the same for each satellite. Overall I doubt that the effect will be serious either for positioning or timing - if it were I think (knowing u-blox) they would make recommendations in the guide. By the way they don't even mention it in the context of their high-precision devices.

(And I've just found an application note where they do discuss matching but only in the context of maximising SNR, i.e. power transfer.)

But if one is really concerned then the obvious thing to do is mount the module/chip right on the back of the antenna so there is no cable.



On 23/11/2016 12:00, time-nuts-requ...@febo.com wrote:
Re: Do reflections up/down the antenna cable cause a problem
       with GPS? (Bob Camp)

_______________________________________________
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Reply via email to