Well, the support case with Cambium is not going well to resolve the LBT on one of our 450 3.65 AP's. The support people just want me to turn down the AP 3db (why not just turn it off?) or put it in a non LBT frequency instead of actually fixing their software that is detecting TDWR radar that does not exist at -62 (spectrums don't see anything else for that matter). Nearest TDWR from the direction of this AP is several hundreds of miles away.
So.. I can't be the only one with this problem. I have tried moving frequencies and power up and down and around, ABAB, not ABAB, channel sizes of all, target recieve adjustments, I've had 3 different firmware versions, no GPS sync errors, no ethernet errors, and LBT stops transmission for 4 seconds and then back on again. Happening at least 2 times a day up to 7 on occasion.
Second question: I haven't been at a customer site when an LBT stops transmission to experience what they would experience. Does anyone know what happens from and end user's point of view here? Would you even notice this as an end user? For instance, would it kick you out of an RDP session or drop a VoIP call? Maybe I'm blowing this out of proportion but I feel like I've invested a big chunk of change for these 450's and I don't feel like Cambium is taking me serious about the LBT issues and they just want me to work around their software glitch that is detecting LBT false positives. At present I only see two ways out of this problem: change country code or change equipment. I don't like fines so...
What other 3.65GHz gear do you have nearby? What's the SM receive target set to on your APs? I use -60. Also, since you have ABAB (four 90's), do you have Adjacent Channel Support enabled on the APs? This must be enabled on all sectors in a cluster if you're trying to run with no guard band between AP carrier frequencies. I have this at many sites and it is definitely needed.
Maybe a competitor? What gear are they using?
Any carrier that hits your 450AP's receiver at -62dBm combined (or -65 on either polarity) will trigger LBT. Decreasing your Tx output power will reduce your LBT threshold. It is a legitimate suggestion. Obviously you get less power down to the SMs though.
Yes, LBT is generally bad for the user experience. Sessions drop, customers get mad. It sucks. But that's the rules of the band.
I have seen LBT hits from our own sectors on adjacent towers. We use PacketFlux devices for sync. These are CMM3 aligned. I discovered an AP running on the on-board GPS. This is CMM4 aligned sync pulse. There is a difference. Enough of a difference for the Tx and Rx from the sites to overlap and cause LBT to trigger.
What software version is this on? It is unlikely that a TDWR station is triggering an LBT event on a 3.65 GHz AP. The onboard spectrum analyzer is not a high performance SA like a standalone SA (.ie. Agilent, Rhode&Schwartz etc.). There is a chance that its missing potentialy instant spikes of interference on channel. That said, do these LBT events go away when you are not doing ABAB frequency reuse? I ask because we have a known issue where when you power up two APs back to back using the same frequency it may cause an LBT event on one or both of them. But this will happen only on initial bootup and not 2 or 7 times a day under normal operation that you are experiencing. Once they are in sync, the APs don't hear each other.
The reason the support team asked you to turn down the power by 3 dB is because this also lowers the detection threshold and can avoid LBT from triggering on this unknown interferer. But I appreciate that this is not desirable as having as much power is available is important for you.
In terms of what happens during an LBT event is very simple. For each instance of LBT event you see, the radio simply skips one frame's worth of transmission. In other words, the radio stops transmitting for 2.5ms and then immediately resumes transmission until the next LBT event. So from a user experience perspective, the outage could be negligible. There could be a blip in a VoIP call or a momentary lag in an RDP session. If you are noticing the radio cease transmssion for 4 seconds, then there really is an interferer that the AP hears causing 4 seconds worth of "silence" from the radio which is around 1600 frames worth of LBT triggers. Under the LBT tab, do you see that many LBT triggers?
Well, George beat me to it while I was typing up the response! Good suggestions from George, there. Thanks!
We are aware of certain 802.11n based non-Cambium equipment that stomp on the 3.65 band, even if they are not operating on the same channel as yours. Even while their SMs are not registered to an AP they cause LBT triggers because they perform active scanning unlike for instance our ePMP product which does passive scanning (and hence are not transmistting unless they are in the process of registering to an AP).