Re: [Discuss-gnuradio] RDF by interference

[Derek Kozel]

Hi Andrew,

You must be able to establish the true time relationships between samples from each radio. The accuracy of the measurement of and stability of the time relationship will be primary contributors to how accurate the calculated direction to a signal will be.

You could use a transmission from a known location as a runtime calibration. By correlating the two sample streams and accounting for the true angle of arrival and geometric relationship of your antennas and the calibration transmitter you can measure the time offsets. Some radios allow a 1 PPS signal or trigger signal to be input and use that to set internal clocks and timestamp samples. This allows for a very accurate initial time synchronization.

A subtlety which is often missed is that if the sample clocks on the two (or more) radios are different then the time synchronization will drift over time. This is why shared frequency references are often used. The combination of the two provides a fixed, stable, and very small time relationship between the two.

The use of a GPSDO in each radio can provide time and frequency references, but both are imperfect. When true shared references are used the imperfection is also shared and so irrelevant for most intents and purpose.

On Fri, Nov 17, 2017 at 1:25 PM, <address@hidden> wrote:

Hello

 

I have a question regarding the use of two software defined radios for the purpose of TOA time of arrival radio direction finding

 

Do the software defined radios have to be locked to each others clocks to work properly or is it good enough that they just pass RF that can be used to determine time of arrival ?

 

Andrew

_______________________________________________
Discuss-gnuradio mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio