Yes, that’s true, but a higher gain antenna at the SM end helps rcv but not 
xmt.  And SM-->AP is the direction you may actually need a better signal 
because the AP likely has a sector antenna and is mounted higher so it sees 
more interference.

 

It would not be unusual to have a 16 dBi antenna at the AP but a 25 dBi antenna 
at the SM.  The antenna gain would help the rcv signal at the SM, but it would 
probably have to lower its conducted power by 9 dB to stay within the 
regulatory EIRP limit.

 

In contrast, in U-NII-3 the CPE end is treated as point-to-point and can use 
antenna gain to exceed the AP limit of 36 dBm EIRP (subject to OOBE limits).

 

From: AF <[email protected]> On Behalf Of Adam Moffett
Sent: Thursday, November 21, 2019 11:27 AM
To: [email protected]
Subject: Re: [AFMUG] dumb DFS questions

 

There might be something I don't understand, but I thought you had flat EIRP 
limit of +30dbm whether it's an SM or an AP.

On 11/21/2019 12:11 PM, castarritt . wrote:

6 dBm loss for the AP transmit isn't the end of the world.  It's the up to 23 
dBm loss on the SM transmit power that destroys the usefulness of DFS for PTMP 
past a couple miles.  The ~16 dBi gain 90° sectors 2-300' up in the air just 
can't hear those SMs over all the noise they are picking up.  What we need is 
the ability to run downlink on DFS and uplink on 5.2 or 5.8.

 

 

On Thu, Nov 21, 2019 at 10:56 AM Adam Moffett <[email protected] 
<mailto:[email protected]> > wrote:

Yeah I think on most equipment you can set alternate channels that are just 
shifted over 5mhz from where you were.  And yeah I think the channel needs to 
be clear for a few minutes before you can go back to it.

Assuming you don't really have a TDWR near you, I don't think DFS events are 
that big of a deal.  My understanding is that DFS events are more likely if you 
lie to the software about antenna gain to cheat the EIRP limit.  False detects 
happen, but I don't think it's a daily event.  Disclaimer: I've mostly used it 
on Point to point with dishes.  I'm not sure if you'd pick up more anomolies on 
a sector antenna.

The biggest bummer is the EIRP limit.  When you're trying to get that 32 SNR 
for the 256QAM then losing 6db kind of hurts.  Or when you've already got 
someone hooked up 10 miles away and lowering the power ruins them.  

Where you really want to use DFS (In my opinion) is at a site where you have a 
bunch of customers within 1-2 miles.  Unfortunately I don't have sites like 
that.

-Adam

 

 

On 11/21/2019 11:31 AM, Ken Hohhof wrote:

We mostly avoid DFS frequencies on APs because of the impact if we get false 
radar detects.  Also we are mostly a Cambium shop.  So I’m a bit confused about 
DFS on other vendor equipment like Ubiquiti as well as home routers.

 

Question 1 – what happens when there’s a DFS detection?  On the Cambium gear, 
we have to select 1 or 2 alternate frequencies.  But on other gear, I don’t see 
this.  When there’s a DFS hit, does it jump to another random frequency?  Does 
it rescan the current frequency until it tests clear and only then resume 
transmission?  Is the answer right in front of me and I’m being stupid?  Maybe 
in the case of routers they are exempt because of low EIRP?

 

Question 2 – what about 40/80/160 MHz channels?  We have a competitor using 
Ubiquiti gear and advertising residential subscriber speed plans up to 100x100. 
 Clearly they must be using at least 40 MHz channels if not 80 MHz, or else 
their marketing people have burning pants and long noses.  And I don’t see how 
a WISP, especially one surrounded by other WISPs, could use wide channels other 
than in DFS bands.  We have some PTP links using 40 MHz but only 10 and 20 MHz 
channels on our APs.  So assuming you are using 40 or 80 MHz in DFS, what 
happens when there’s a DFS detect?  Does the whole 40 or 80 MHz have to find a 
new home?  Can it slide over 2.5 or 5 MHz and substantially overlap the 
previous occupied spectrum?  DFS bands come with enough spectrum to use wide 
channels, but is there enough to jump around when you take a DFS hit?





-- 
AF mailing list
[email protected] <mailto:[email protected]> 
http://af.afmug.com/mailman/listinfo/af_af.afmug.com





-- 
AF mailing list
[email protected]
http://af.afmug.com/mailman/listinfo/af_af.afmug.com

Reply via email to