We have a large WISP network where we are running a number of licenced, and class-licenced point to point links.
We have had this problem a few times in the past and can't quite seem to work out the root cause just yet. We would like to see if any fellow WISP network operators can provide their perspective on this.
The latest incident we had included 5x of our PTP links (4 licenced, 1 class-licenced). What we see is the tx and rx levels drop over a period of between 10 minutes and 2 hours, and consequently the tx/rx levels become so weak that the links drop. All 5 of our links went down for between 2.5-3 hours (depending on the link). Then, the links sharply recover to their normal operating tx/rx levels over the course of 30 minutes.
This would initially indicate to us that there is weather in the area affecting the links, but there was no storm or rainfall in the area on this day and time. The links are spread out over 7 different tower locations. The towers are within a 80km radius. The links are all on different channels if on the same frequency, or on a totally different frequency.
Has anyone else experienced a similar problem before?
You'll need to provide the forum with much more detail to get a meaningful response... for instance, what is a 'class-licensed link'? What frequencies are all these PtP's operating at? What equipment are you using for each PtP that is effected? Can you describe what kind of weather you're experiencing when this issue occurs? Have you filed a detailed support ticket directly with Cambium yet? And my favorite question of all... are you running the newest firmware release on all radios?
Haha, class licensed must be an Australian term only. Class licensed is 5GHz and 24GHz. The links affected are 7,8,11,18 and 5ghz although it doesn’t matter what frequency they are as they all experienced the exact same problem and the range of frequencies affected varies alot. As I mentioned before there was no rain or storm. It was a fine sunny day. And no, I don’t plan to open a detailed support ticket this query doesn’t belong to them, it’s not specific to just our cambium equipment. BTW, there is a mix of cambium equipment and other vendor equipment. Firmware is not relevant in this case given that radios from 3 different manufacturers all joined in at the same time to enjoy the outage party.
The key question I have is - has anyone heard of this or has had a similar experience?
High and low latitudes (such as australia and sweden) are more affected than equatorial regions, so that may be why not everybody sees such strong anomalies.
To me this sounds like refraction, due to an inversion layer. An inversion can ocur where there is a significant difference in the air temperature at different heights see here for more detail https://en.wikipedia.org/wiki/Anomalous_propagation
I have seen it here where two high sites link over a lower plain, and there is fog on the lower plain, which means the air is cooler and therefore more dense than the warmer upper air. What you can do is set up a divesity system, with dishes vertically separated by a few metres, statistically the fade will not occur at both places at the same time.
You can also see a similar effect over tidal water, although the cause is not the same.