In my career I’ve made mistakes and I’ve screwed things up.
But I hope I’ve never done anything as monumentally clueless as the following I received in an email from a customer’s “RF Expert”.
I’m not disclosing the name or affiliation, it’s just too embarrassing for them.
“If your mod counters are correct, and flying all over the place and your SNR values are good on both sides, then it really appears to be a problem that we have seen in the past with “mid-path” interference. Since your SNR values are good on both sides, that means the localized noise floor at the AU and SU is low (noise-RSL=SNR). However, if there is a “mid path” interferer, the packets will get corrupted as they traverse thru the path. (see diagram).
We have seen this about 10 times in the past, primarily with PTP links, where both side show great signal, spec scans look fine, but the link either has BERs or the data rates (in Adaptive mod) are all over the place. Depending on the amount of usage on the PTP interferer, the degree of corruption will vary as the link may not be transmitting all the time.
You can run spec-ans at the SU and AU all day but you wont see the noise if this is the case. It also may explain why some SUs are fine as they may not be shooting thru the interferer depending on angle and distance.”
And yes, he included a helpful diagram:
I know, I’m stunned too. It’s amazing.
I printed the email and pinned it to the wall so I can look at it when I’m feeling stupid.