I will mention I do not have this issue with my Tarana 6ghz SMs. However they do not show me the MSL height so I am not sure what they send to the AFC system to calculate allowed EIRP.
Maybe the Cambium GPS puck and setup is just not really that accurate?
To help track this GPS height problem I created an SNMP probe to graph the SM height polled every minute.
Here is the SM reported height for the last 2 hours. As you can see it varies at times quite a huge amount.
The AP which is higher than the SM by around 100 ft does not have any level close to this variation. It also has better GPS signal tracking, but also uses a different GPS chip and puck.
I will also note that I am in flat Indiana farm land with no hills or mountains. Just farm land and then some groups of woods or tree rows. No woods or tree lines within 1000ft or more of this SM.
Based on the logs you shared and verifying allowable EIRP at this SM location on 6GHz Spectrum Availability tool, I came to the same conclusion as you did - that AFC response resulting in different EIRP is the result of changing heights reported by the GPS. With the GPS puck on 4625, it’s expected to have +/- 50 meters variation in altitude/height measurement. It looks like, at this lat/long at higher altitudes, 6GHz licensed microwave is nearby and this results in AFC response of lower EIRP. With that said, this issue may not be present on other SMs, as an example, SMs at some other lat/long for which even higher altitude points do not have licensed microwave in the vicinity, will not get lower EIRP.
With all that said, based on this issue on this SM, we plan to do software improvements to get rid of this issue.
Thank you for looking into it. I would love some type of software fix. Not exactly sure how you do it as I don’t want people to be able to cheat on the height, but at the same time a +/- 50meters accuracy is quite large and as you saw can be a make or break for certain locations.
Not sure if you can do the MSL height using an average or mean of what the GPS receiver reports over time to try to lessen variation, or maybe keep track of the shortest GPS height and always report that to the AFC? Another option could be to take the GPS reported MSL height don’t let the change go higher than the operator defined AGL height, but limit it to + height changes only. If the operator put in a negative AGL then ignore that value and just use the GPS reported height.
I’m sure there are other ideas from much smarter people than me, but again +/- 50 meter accurracy just might not be accurate enough from some locations.
A possible fix might be to manually enter in the height of the SM and then use GPS to verify the entered height. If the two values are within like 10-20m, then the height value submitted to the AFC is locked, and going forward, large deviations in height would be ignored.
I have a small update on Tx Power limit issues related to GPS. The current status is that we are aware of it and our engineer is assigned to get some solution. He is gathering and structuring information.It is not clear yet how it should be. It will be discussed with all stakeholders, including FCC, Qualcomm and our development team.
Also we will add a GUI button to re-query AFC server for debug purpose.
Thank you for the update . I was just going to chime in: I have an SM that is also experiencing the same issue as the OP. GPS is magnetically attached to top of dish like OP has done and has fairly clear view of sky but my GPS signal can waiver and i suspect when it does the SM height changes and gets reduced, sometimes even below standard power (i saw it once at -1dbm TX power). Subsequently we have to hold off on any fixed client installations and only rely on standard power at the moment. My case is a bit stranger because my AFC status shows im being approved for 36dbm on this SM for the entire U-NII-7 band but the SM does not set itself accordingly. Maybe I am experiencing a different issue?
EDIT: to add: if I use the AFC tool in link planner and set this SM to 1000ft in the air (when it is only mounted at 50ft) the planner shows I should still have full 36dbm (i am fairly remote) yet TX power still fluctuates from time to time
I have been charting the GPS height for the SM and the AP for the last 3 days and you can see both the AP and SM have quite the variation in height at times. SM seems to vary more as the GPS signal is not as good as the AP, but the AP has a great GPS signal and it can still vary in height be over 100ft.
As mentioned before that height difference would not be an issue if the GPS was just used for TDD sync but since the AFC takes the equipment height and uses that to determine power levels having an accurate and stable GPS is very important.
SM height
AP height
5 posts were split to a new topic: Mounting option on ePMP 4600 SM
