In summary, the data that I received from TTN seems a bit odd to me. Of course I could imagine explanations for all the odd cases I encountered but to my eye the data just does not look 100% right. This might be due to a bug hidden in my code or TTN or (and more likely) due to my lack of experience with Lora signal propagation in urban environments.
For what I understand, the latter is key; you’re (at least) running into reflection/multipath problems?
Even with static nodes and antennas you can see RSSI variations of 2dBm perhaps 3dBm. If the node is moving along a street (a car say) then signal strength variations of 10dBm over a couple of metres are again not unusual due to the impact of changing reflections etc.
Now consider what these variations men for a node that is 1km from a gateway. A +2dBm change in RSSI means the node has apparently moved position by 400m. A +10dBm chage in RSSI means the node has apparently moved postion by 3km.
And perhaps dont run a node sending a packet every 20 seconds for very long, you can breach the fair use limit fairly quickly, about 20 minutes at SF10 BW125.
Thank you for sharing your interpretation of the effects. Maybe one additional question: How would you interpret the situation where all close-by gateways received nothing and only the ones very far picked up the signal? I think it was in the order of 5 GWs getting nothing and two at a (urban) distance of 8km received something.
Not much to be gained by thinking about complex situations.
Reduce the test, if there was one, to the simplest situation, otherwise you will probably learn zilch.
You have one node, does it really only get picked up when its 8km from a particular gateway, yet it does not get picked up when its say 100m away from the same gateway ?. Is the same true for all of the other gateways ?