FrSky RX antenna diversity (XSR)

[Carbo]

There is not much information out there, about FrSky RX antenna diversity. So i did some investigation and want to share the findings. Frontend in the XSR is a Skyworks SE2431L, with two antenna inputs and an antenna selection pin.

If the RSSI is over about 44dB only one antenna is active. Otherwise the RX switches between the antennas until the RSSI is again over 44dB. The two 60s logs, RSSI and antenna select GPIO, are roughly synchronized.

The RX switches every 4 frames at low signal level:

[jhsa]

It looks like it might not have anyhting to do with RSSI.
It just switches antennas if 4 packets were lost.. That is what it looks like to me.. More or less the same as the D8R-II Plus..

João

[Carbo]

Of course it is possible, that the switching happens after 4 lost frames, do you have informations or measurements about this? Frameloss begins at about 44dB RSSI. Here is a G-RX8 rangestest, where the frames are also logged 100=100% successful received frames. This shows, that there is a relationship between RSSI and frameloss. The user has usually only the RSSI information, to judge the reception quality.

[jhsa]

Not always, you can start losing frames at high RSSI. Swamping for example, where the signal is too strong.
I can actually make my DIY RX start switching antennas at will, by swamping it with high signal..

I just did a little video showing this.

https://www.youtube.com/watch?v=sB2TgnIQ7lw

The RX starts losing packets and it will switch to the other antenna, until it starts losing packets again. This is simple to achieve because you don't need to know which antenna is receiving what, you can even swap the control pins on the switch. It is a very simple circuit. If packets are lost, it will just switch to the other antenna, if it continues to lose them, it will switch quickly between both of them.. I also tested that today with a servo test programmed on my tx, and the tx in range test mode. Then moved the rx away until I started to get quite a bad signal,and watched the LEDs. When they were flickering backwards and forwards (losing lots of packets), the servo was still moving. When I covered one of the antennas completely, servo stopped..

When Working (doing testing) on the DIY Frsky compatible receiver, we did connect some LEDs to the D8R-II Plus switch control lines and watched what was happening.. We tested diversity based on RSSI as well but the result wasn't as good.

João