We’ve been installing our network of stations in Belgium (Project #959) and we are troubleshooting some problems that we’ve been experiencing. I could use some help!
Two of our stations (#12048 NP Bosland - Ecoduct Waaltjesbos, and #12018 NP Bosland - Heusden terril) seem to be properly connected to a cellular network and send data from our test tags through. However, when I look at the deployment timeline on the Motus website, the activity on the antennas seems to be fragmented. Is this normal, or is this indicative of bad connections in the antennas/coax cables?
One of our other stations (#12072 Zwarte Berg) seems to be not connecting to a cellular network. The blue light from the modem is blinking, but the red diagnostic light does not turn on. Our phones show that connection to a cellular network should be rather easy at that location, and mails with CTT confirmed that the SIMs are active and functioning properly. A test with our remaining receivers (not deployed yet), showed that one other receiver had a similar problem. Is there any other reason why the sensorstation would not be connecting to the cellular network?
To clarify, we use SensorStations V3.2 in all our stations.
That antenna activity looks very normal to me, and tends to vary station-to-station and even between antennas based on local ambient noise.
When diagnosing the cellular issue, did you go to the cellular carrier info page on the station LCD screen? That page allows you to see the carrier that the station is attempting to connect to and the signal strength. Perhaps it is connecting to the tower but doesn’t have a strong enough signal, and an external, larger cellular antenna could solve your issue.
As Nick says, the antenna activity on those timeline really does vary, and gaps aren’t unusual. And yes, it can also be due to bad connections or faulty antennas.
Something does look a bit odd with those though. The GPS bar is unbroken, so something is being transmitted over the cell network. All 4 FCD turn off right at the time the “reboot odomter” increments. Following that, several FCD do come back online, often at the same time, before cutting out again often at the same time.
This may be related to a very long-standing issue related to how multiple FunCubes are handled by some versions of SensorStations and Raspberry Pi-based SensorGnomes. The issue has been fixed by CTT in later versions of the SS software image, though there remains an inherent challenge in running more than 3 FCD while also running the cell modem and WifI module. The newer SS software provides toggles to turn the cell and / wifi off if needed. You can read more about that at the link below. Hopefully the steps here resolve your issue!
I will visit the station in the coming days to check the cellular carrier info page!
With regards to the issue with handling multiple dongles. We use Nooelec NESDR Smart dongles. So, I will go and disable the wifi module, since we’re not using it, but according to the link you sent, it should be less of a problem with non-funcubes?
We’ve also been having some problems with tag detections. We have several nightjars with a Motus tags that should fly around within the antenna detection range (verified with a handheld antenna), but there are no detections coming through. Now, I know that the Nooelec dongles are not as strong as the funcubes in detecting these signals, but since we’re within 2 km of the station, it’s a bit strange to me that even no single, short, sporadic signals are coming through.
Moreover, the last two stations that we’ve tested with test tags did not send through any of the detected test signals. When looking on the sensorgnome interface, I could see the two test tags being detected on different antennas, but when downloading the data through R, I cannot see any signal on these stations.
Can this be related to the problem of the antenna activity?
My understanding is that, yes, this issue is more relevant to FunCube Dongles since they are quite old and use a very outdated protocol (USB 1.1 if I recall). Someone like @TvE is probably in a better position to comment on the potential comparative impact. Either way, since you’re not using the WiFi, there’s no harm in disabling it.
The coloured bars on the deployment timeline that you copied above represent hours in which any radio activity on a Lotek frequency was recorded – whether or not it resolved to something like a “hit”. The pattern in the timeline appears to show the dongles failing completely and recording nothing whatsoever. If your handheld detections occurred during those periods, it would make sense that you didn’t get any data.
But if, as you suggest, you also saw those tags on the SG interface, there maybe something else going on, and it may be related to the important differences between the standalone tagfinder that runs on the SG and the global tagfinder that runs on the Motus server. There is a little bit about that here. And there are multiple filters that the processed data goes through which might also affect what you’re seeing. Though the complete processed data is always available in R so you should see everything, even runs of length 2.
Hello Jitse. I just checked on your station and it looks like things are running well. You might want to try one of the Power Filters I just posted about in the forum which may increase detections at your station. It looks like your power dropped over the course of several days but is now recovering. I’m not sure how you are powering your station, or if your solar panel went down at some point, but right now everything is looking good.
