Thought I’d pick up this thread again because I’m looking hard at CBRS
LTE for my densely forested town, largely because of its alleged
foliage penetration.
What’s anybody understand the EIRP limit for a 20Mhz channel to be now
in CBRS 3.65? Can I assume that modeling RSSI in a tool like RMD can
serve as a rough equivalent of RSRP? Vendor is telling me that where
he heatmaps a -100dBm signal represents full modulation—does that make
any sense? Maybe he’s being a little slimy and referring to uplink
modulation on a 1T4R UE?
And David—you started this thread and said you were trialling those
various platforms—anything to report? Did you get your hands on the
Baicells and/or Airspan stuff?
*From:* AF <[email protected]> *On Behalf Of *Adam Moffett
*Sent:* Monday, September 14, 2020 8:50 AM
*To:* [email protected]
*Subject:* Re: [AFMUG] LTE vendors
For CBRS, depending on antenna and channel size, yes it's probably
legal. When I went to that Telrad training session a few years ago,
CBRS was still a hypothetical thing and everyone there was operating
under an NN license with the 1W/Mhz EIRP limit.
And yeah that's how ALL wireless works. At the moment in time when
the AP is talking to a station at 1Mbps, the capacity of the channel
is 1Mbps. At the moment in time when the AP is talking to a station
at 300Mbps, the capacity is 300Mbps. The average capacity over time
is going to be a function of how much time is spent talking to each
station at each rate. If you literally had one at 1Mbps and one at
300Mbps and both were allocated equal airtime then your capacity would
be 150.5Mbps. It's true that a 5Mbps UE won't make the capacity of
the eNB 5Mbps, but it is true that while the channel is being used to
talk to that UE, the channel is only running at 5Mbps. My point was,
if someone is testing with a single UE and happy that they're getting
5Mbps, then they're forgetting that they won't actually get 5Mbps when
there are other UE operating at the same time, and that the weak
connections they install are weakening efficiency of the whole
sector. I know you know this, I think you're just misinterpreting
what I said.
On 9/14/2020 8:39 AM, Matt Hoppes wrote:
Hold on. 30dBm is well within legal power for CBRS.
Also a station connected getting 5 megabits is not dragging the
entire sector down to 5 megabits. That’s not how LTE works.
On Sep 14, 2020, at 8:34 AM, Adam Moffett
<[email protected]> <mailto:[email protected]> wrote:
Attenuation in 3.5ghz is on average 15db per 100meters of
foliage. I got that number from a Telrad engineer, and it
seemed to hold up experimentally. Whether it's Wimax, LTE,
etc, there's no reason that would be different.
LTE can connect with almost nothing for a signal. So a person
testing with a single base station and a single UE might run
around and say "wow I've got 5 megs here and No LOS!", but I
think they forget that the entire base station's capacity is
5meg when it's talking to that single UE at 5mbps. It's
impressive that it worked, but is that actually useful as a
fixed ISP?
Another thing I noticed is that Telrad could turn the Tx Power
all the way to +30dbm, and people were actually doing it, and
Telrad support seemed to be encouraging them to do it. At a
training session someone in Telrad support told me, "Adam, if
you're worried about the legal EIRP limit then you're the only
one worried about it." So if you're 8-10db stronger than the
legally operating product, and you can technically connect
with a signal too weak for the other product, that certainly
makes people feel like there's better penetration.
There may also be some "magic" in how LTE allocates resource
blocks and gets feedback from the UE's (CQI) on which resource
blocks are working best for each unit, but I think that's a
matter of getting the most value possible out of a trashy
signal. If you're a fixed operator building for capacity and
performance then you hopefully won't be installing with a
trashy signal anyway.
My biggest issue of all is that all of the WISP priced LTE
stuff is clunky and buggy. Frankly, that was true of WiMax
too. It seemed like Telrad's bridging modes never quite
worked right for example. You were better off building an L2
tunnel on your own box behind the UE.
-Adam
On 9/14/2020 12:19 AM, Ken Hohhof wrote:
Ever since I got bamboozled into deploying a WiMax
basestation, I have been skeptical of tree penetration hype.
We have been deploying Cambium 450 in 3.5 GHz / CBRS and
it’s great, but it doesn’t “penetrate” trees. OK, an SM
within a mile can go through 1 or 2 trees, depending on
the size/density/type of tree. And with the usual caveat
that trees near the customer are more problematic than
trees in the middle of the path.
Some people say otherwise, but there were all sorts of
glowing testimonials for the WiMax equipment as well.
Maybe LTE has magic properties. I doubt it, but I haven’t
tried it, I don’t want to repeat the WiMax fiasco. So I
could be wrong. But when I’m wrong, usually it’s because
I wasn’t pessimistic enough and things are even worse than
I feared. Only on rare occasions do I expect a lion
behind the door and there’s a beautiful lady. Usually
there’s 2 lions.
Certainly turning on CBRS made all our 3.5 GHz Cambium
stuff work better, we got several dB higher xmt power, and
usually cleaner spectrum. But the cleaner spectrum thing
is only true until other operators fire up their stuff in
3550-3650. Even if you get a PAL, it’s not like nobody
can use that frequency in the whole county. The
interference at the edge of your PAL protection zone
should be below some level that the SAS uses when
authorizing nearby operators to transmit. But that level
isn’t -99 dBm.
LTE gear may be designed with better receiver sensitivity,
that will help if the noise floor is really really low.
On the other hand, does most LTE gear use the highest
allowed EIRP? What about the CPE? That was another
problem with the WiMax stuff, the CPE was 3rd party stuff
that typically had kind of wimpy xmt power and not
particularly high antenna gain. Maybe that’s not true of
LTE gear, I haven’t looked into it. But pull out a
Cambium 3 GHz 450b high-gain SM spec sheet and compare to
the LTE CPE.
*From:* AF <[email protected]>
<mailto:[email protected]> *On Behalf Of *Trey
Scarborough
*Sent:* Sunday, September 13, 2020 4:43 PM
*To:* [email protected] <mailto:[email protected]>
*Subject:* Re: [AFMUG] LTE vendors
Has anyone done a comparison or know of a whitepaper
between LTE and Cambium? I am mainly looking at tree
penetration or lower DB signals to actual throughput
comparison. I have been told that LTE gets a little better
tree penetration but if that is at a low rate that really
doesn't help any.
On 9/12/2020 10:03 AM, Darin Steffl wrote:
It comes down to complexity. Ericsson, Nokia, etc are
all cellular brands and to run and manage those
complex LTE networks, you need full time engineers to
manage, debug, and optimize things.
Cambium is so easy, in comparison, there's very little
extra learning to do in order to get it running great.
Ericsson LTE probably would require months of training
and needing to hire someone just to run the gear or
hire expensive consultants to do it for you.
On Sat, Sep 12, 2020, 9:49 AM Kurt Fankhauser
<[email protected]
<mailto:[email protected]>> wrote:
450m is the only way to do, especially if your
already using the 450 platform in other parts of
your network, there is an operator in my area with
the Ericson system and they had a ton of issues
with getting it up and running, not even sure if
they ever got it all resolved.
On Fri, Sep 11, 2020 at 9:00 PM Sean Heskett
<[email protected] <mailto:[email protected]>> wrote:
Yup what josh said lol.
We tried the LTE thing and glad we switch to
450m...much easier.
-Sean
On Fri, Sep 11, 2020 at 4:43 PM Josh Luthman
<[email protected]
<mailto:[email protected]>> wrote:
Having done one LTE vendor and 450m the
only mistake I made was not buying the
450m sooner.
Josh Luthman
24/7 Help Desk: 937-552-2340
Direct: 937-552-2343
1100 Wayne St
<https://www.google.com/maps/search/1100+Wayne+St+Suite+1337+Troy,+OH+45373?entry=gmail&source=g>
Suite 1337
<https://www.google.com/maps/search/1100+Wayne+St+Suite+1337+Troy,+OH+45373?entry=gmail&source=g>
Troy, OH 45373
<https://www.google.com/maps/search/1100+Wayne+St+Suite+1337+Troy,+OH+45373?entry=gmail&source=g>
On Fri, Sep 11, 2020 at 5:54 PM Adam
Moffett <[email protected]
<mailto:[email protected]>> wrote:
And yeah, 450m might be expensive, but
so is all the LTE stuff.
You'll max out the legal EIRP with
450m, and get 8x8 MIMO. I think
part of the magic with LTE is that it
will connect with ridiculously
low signal, but on a fixed system you
probably won't really want the
trashy signals anyway.
Cambium also has LTE for whatever it's
worth. The CBRS version
is supposed to be available relatively
soon (though I forget
precisely when).
I don't know if I state it as "fewer
issues since there is no
EPC", but definitely fewer
complexities and fewer things to worry
about. The connection from eNB to EPC
has to be /pristine/,
and the EPC comes with its own set of
new terminology and new
concepts to figure out.
On 9/11/2020 4:06 PM, Darin Steffl
wrote:
I have seen lots to people doing
450M in CBRS
stating coverage is nearly the
same as LTE but way better speeds
and triple the aggregate capacity
due to mu-mimo.
Way fewer issues too since there
is no EPC. Just
straight layer 2 with no bullshit.
On Fri, Sep 11, 2020, 2:39 PM
David Coudron
<[email protected]
<mailto:[email protected]>>
wrote:
We are looking at a new area to
expand out network that has a
lot more tree cover than
our current footprint. We
are thinking with the
combination of CBRS and LTE,
that we might be able to
offer better coverage than
with traditional fixed
wireless options. We have
started conversations with
the following vendors,
wondering if anyone has any hands
on experience with any of them
and what their
impressions were:
Blinq
Airspan
Baicells
Ericsson
The Ericsson equipment is in a
class
by itself price wise, but the
others are similarly
priced, and somewhere around
double the price of PMP 450
stuff. Normally we would add
more tower sites for
better coverage, but this
project will need to be done
before the end of the year and
building towers isn’t an
option. We have good enough
spread on the towers that
we think we can do this with
PMP 450 APs, but are
thinking we’d get even better
coverage out of LTE. Any
opinions on the reliability
and the manageability of the
four vendors above? Sorry for
such an open ended
question, but not sure what to
ask to be more
specific. We know that we
will have the LTE stuff to
deal with like access to an
EPC and so on, so not so
much worried about that as
more the manufacturers
themselves. Baicells
concerns us as they may get
lumped in with Huawei.
Thoughts?
Regards,
David Coudron
--
AF mailing list
[email protected]
<mailto:[email protected]>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
<http://af.afmug.com/mailman/listinfo/af_af.afmug.com>
--
AF mailing list
[email protected] <mailto:[email protected]>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
<http://af.afmug.com/mailman/listinfo/af_af.afmug.com>
--
AF mailing list
[email protected] <mailto:[email protected]>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
<http://af.afmug.com/mailman/listinfo/af_af.afmug.com>
--
AF mailing list
[email protected] <mailto:[email protected]>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
<http://af.afmug.com/mailman/listinfo/af_af.afmug.com>
--
AF mailing list
[email protected] <mailto:[email protected]>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
<http://af.afmug.com/mailman/listinfo/af_af.afmug.com>
--
Trey Scarborough
VP Engineering
3DS Communications LLC
p:9729741539
--
AF mailing list
[email protected] <mailto:[email protected]>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
<http://af.afmug.com/mailman/listinfo/af_af.afmug.com>
