The link: 4.5 miles, 1 Big fat building in the way, barely unable to clear
the roof. Noise floor high.
Limits: Noise Floor to high for PtMP Trango, based on obstruction.
Stats: rssi -75 & -78, noise -79 or worse on Horiz, Vert worse, RSSI almost
15db below calculations due to NLOS )
Solution: Install PTP to get more gain on AP side, Add OFDM to help with
obstruction.
Trango 5830 was invaluable to determine what was going on. It's built-in
survey command was able to determine the noise floor on all channels
accurately, and home in on the fact that the link was marginal because of
gear that used a 20Mhz channel half way between Trango's channels.
StarOS w/ 28 dbi PAcwireless on both sides- Got -55 & -60 rssi. Good link,
but it was not perfect, with 1 out of 20 large ping packets with high
latency. It would regularly negotiate down to 36mbps or 18 mbps on one side.
StarOS w/ 28db on one side, and 23dbi on other side- Got -60 & -65 rssi.
Excellent / Perfect link. Stayed constant at 54 mbps, with a very rare
negotiation down to 48mbps or 36mbps. We believe this is becaue one of two
reasons, reflections off the building right back at us, or the wide
beamwidth of lower gain antenna to help use multi-path to optimize OFDM. We
often felt 19-23 dbi antenna ideal for OFDM. This put us above the noise of
most of the channels, and narrowed our beam compared to PtMP to reduce
noise. OFDM clearly helped to not lose rssi due to the building
obstruction, and gain was not received solely from higher gain of PTP
antennas.
The problem with STAROS-V3... We ran survey, and picked up ZERO interference
or devices, but yet we know that there is lots of interfering devices out
there. The "Quality" reading was pointless at either 100% or 13% with very
little correlation to what the link actual performance was. Hard setting
modulation, to 24mbps, left the link unusable, even when Quality of 100 was
shown. When we put modulation on auto, every thing worked well. SNR was
only available on client side, and not accurate, reading only a -95 (which
may have been average, but not peak noise, based on Trango scans).
Basically, with the STAROS box, we were left totally in the dark, on what
the noise environment was. We really missed the detail of the Trango tools,
and not sure what we would have done, if we had not had a Trango on site
simultaneously gathering test results. We learned via the Trango, that we
could have survived the noise with a 10 Mhz channel, that the StarOS
allowed, but we would not have known where that was without the Trango test
results. We relied on End to End large pings to determine link state during
tests, and were glad to see the addition of Iperf embedded in StarOS for
more strenuous testing afterwords.
The end result... We left the StarOS installed for a perfect link, and
defined many possible options should interference need to be battled in the
future. We saved a bunch on hardware, costing us under $1000 in equipment
for the link, and delivered the highest quality link, as any gear could
offer.
But this brings me to my point of this post. What was the true cost of this
job? I spent a day installing Trango PTMP. I spent a day isntalling StarOS,
both with two engineers. I lost a months revenue, delaying my trips between
upgrades and tests.
At a price, All these headaches could have been avoided. Most likely Trango
Atlas PTP would have solved the problem and given us the benefits of Trango
testing tools, and OFDM, and price under $3000. But there was some risk in
trying that solution. In the past we've had difficulty in high noise
environments, and/or to high of RSSI. We did not have an Atlas on hand to
test.
We took the time to do a test with Alvarion B40 that we had on hand. The
Alvarion picked up the noise in its survey. The Alvarion gave us accurate
SNR readings that we could use to best plan the link configuration. And the
link quality was perfect as well using the 28dbi and 23 dbi antennas. So
had I used the Alvarion VL to begin with, I would have saved our company two
days in labor, and would have had all the tools that I needed to install the
link easilly the first time and to adapt in the future. Alvarion clearly
would have been the winning choice. It gave me confidence that in future
jobs IF I had to design a link in advance blind, I could order an Alvarion,
and it likely would best be qualified to complete the job successfuly.
I ended up keeping the StarOS in place. The reason was two fold. 1) I
already spent the time, so why not save the money on equipment. And
secondly, at the AP side, I wanted to add a second radio card. Because I
switched the link to PTP, the other client that was being served via the
PtMP, still needed to be served. For $100, I was able to add the second
card, and install a second sector to serve that subscriber still. (two
sectors for the price of one).
Every product has its value. You be the judge on what product will best suit
your next project.
Tom DeReggi
RapidDSL & Wireless, Inc
IntAirNet- Fixed Wireless Broadband
----- Original Message -----
From: <[EMAIL PROTECTED]>
To: <wireless@wispa.org>
Sent: Tuesday, September 26, 2006 7:48 PM
Subject: RE: [WISPA] vendor specs -- Jon
Patrick, ditto on the 3650 band. However the reality is that self and
external
interference in the UL world is all too common. You say UL bands or at
least VL doesn't need GPS capability because of so much capacity. If
you want I can get you a list of wifi/trango/etc.-to-Canopy 'converts'
that will tell you otherwise.
Licensed carriers use GPS to greatly diminish what we experience as common
day
interference problems. IMO I can't blame the FCC for not giving more
spectrum than they have as we've already trashed what we've been given.
Lastly, what Moto did was brought GPS sync to the UL world however as
standard option and in very economical form factor, not expensive
chassis and such. If you haven't already, get your VL guys with your
WIMAX guys and you could have a clear winner down the road! :)
Jon Langeler
Michwave Tech.
Quoting Patrick Leary <[EMAIL PROTECTED]>:
Jon,
For sure I'm all over GPS for all licensed (world of small channels) and
when there is a small amount of spectrum to work with in UL. For
example, in the coming 3650MHz band, GPS should be a must for PMP. Same
with scaled 900 (we offer it there). It is just not needed with VL. What
for? It already gives massive capacity without any re-use. Even with GPS
and re-use I do not think Canopy can get close to the amount of capacity
VL can offer. Frankly, even if we had it for VL no one would buy it.
No argument from me on the scheduled MAC front, except to the extent
that in UL it needs to come with good interference mitigation (not
talking about self-inflicted interference) techniques to make it useful.
Patrick
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of Jon Langeler
Sent: Monday, September 25, 2006 10:37 PM
To: WISPA General List
Subject: Re: [WISPA] vendor specs -- Jon
Hey Patrick, GPS...there's many reasons and it's not a canopy vs
alvarion debate from my standpoint, more so a scheduled mac(canopy,
wimax, 3G...) vs unscheduled(wifi, VL, currently Trango). I'd predict
that as wisp education progresses, they will realize the power of
scheduled mac and GPS support. By then maybe the rest of the BreezeMAX
code will have made way to the VL engineers and everyone can be happy
:-)
Jon Langeler
Michwave Tech.
Patrick Leary wrote:
Jon,
Why is that the case? You really think GPS on Canopy is some cool
feature? Canopy must have GPS to function. Without it, it kills itself.
It is all to prevent self-inflicted interference (remember, Canopy does
not even have ATPC) and to allow for channel re-use. Other systems,
like
VL, do not need it. It provides far more capacity than Canopy, so it
does not need to re-use channels and with basic channel planning you
don't have issues with self-interference.
Patrick Leary
AVP WISP Markets
Alvarion, Inc.
o: 650.314.2628
c: 760.580.0080
Vonage: 650.641.1243
--
WISPA Wireless List: wireless@wispa.org
Subscribe/Unsubscribe:
http://lists.wispa.org/mailman/listinfo/wireless
Archives: http://lists.wispa.org/pipermail/wireless/
************************************************************************
************
This footnote confirms that this email message has been scanned by
PineApp Mail-SeCure for the presence of malicious code, vandals &
computer viruses(192).
************************************************************************
************
************************************************************************
************
This footnote confirms that this email message has been scanned by
PineApp Mail-SeCure for the presence of malicious code, vandals &
computer viruses(42).
************************************************************************
************
************************************************************************************
This footnote confirms that this email message has been scanned by
PineApp Mail-SeCure for the presence of malicious code, vandals &
computer viruses.
************************************************************************************
--
WISPA Wireless List: wireless@wispa.org
Subscribe/Unsubscribe:
http://lists.wispa.org/mailman/listinfo/wireless
Archives: http://lists.wispa.org/pipermail/wireless/
--
WISPA Wireless List: wireless@wispa.org
Subscribe/Unsubscribe:
http://lists.wispa.org/mailman/listinfo/wireless
Archives: http://lists.wispa.org/pipermail/wireless/
--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.405 / Virus Database: 268.12.8/455 - Release Date: 9/22/2006
--
WISPA Wireless List: wireless@wispa.org
Subscribe/Unsubscribe:
http://lists.wispa.org/mailman/listinfo/wireless
Archives: http://lists.wispa.org/pipermail/wireless/
