Well I have had 2.4ghz radio's link up at -89db (not very well mind you
but...) so I don't know what to tell you other then Moto has
traditionally understated there spec sheets. The GPS is what sets the
timing for the AP's. The AP's coordinate the timing slots for all SM's
registered to them. So how it works is that all AP's on channel 1
across the world all transmit at the same time, and all SM's synced to a
AP on channel 1 with GPS timing from the AP listen at the same time.
Distance is not relevant unless you are utilizing the feature set of the
SM to retransmit a GPS sync pulse that it receives from and AP to a BH
or AP. The lag that is introduced by having to transmit that pulse info
across the wireless link to the SM retransmitting is the only time that
distance can come into play. The application this is used for is for a
cheap repeater system so that you dont have to have a GPS synchronizing
device at every tower.
/>SM
GPS -->AP#1 /
\
\>SM (retransmitting GPS sync pulse) -->AP#2
-->SM (retransmitting GPS sync pulse) -->AP#3 (this AP will be out of
sync with AP#1)
Basically the timing is measured in nano seconds so it takes to long for
RF to transmit the data across the wireless links to continue to
propagate the timing signal. But if you put a GPS sync generating
device at AP#3 it would be in perfect time with AP#1 and close enough
timing with AP#2 that they all would get along.
One thing to keep in mind is if you are the only Canopy shop in the area
you can have your AP's generate the sync pulse and avoid the cost of the
GPS synchronizing items. Also again as for the distance statement. 6
AP's in a cluster sharing 3 channels have to be synced. believe me the
messy antenna on the Canopy units dont have a good enough F/B ratio to
not hear another AP 6" away from it. The two AP's that are back to back
share the same channel so that when they transmit the SM's that are
listening are as far away from each other as possible and thus reduce
any chance of talking over each other. The largest benefit that GPS
sync allows is to add additional capacity to area's by allowing for more
towers to be in a smaller area without self interference. If long range
rural deployments are the plan then GPS sync will only benefit you if
you have competitors utilizing the same equipment and configuration in
the area. So a Moto advantage cluster has about 84mb total (Classic
Canopy would be 42mb) FTP bandwidth available to it. If more is needed
you can place the towers with in a few miles and divide a cell into two
micro cells each with a possible 84mb of total bandwidth for a total of
168mb serviced to a given area.
One last note, GPS timing will not allow for two separate clusters of
the same type ( two 2.4ghz clusters) to be on the same tower. I can't
write out whats in my head on this.... getting a little late in the
night but if you wanted to I could talk to you over the phone and
explain it. Send me an email to anthonyw (at) broadband-mn.com and Ill
give you my cell phone number or give you a call.
Anthony Will
Broadband Corp.
Travis Johnson wrote:
Hi,
First, the spec sheet on Motorola's website says -86 RSSI.
What happens when you have more than 3 towers outside of the 8 mile
range of GPS sync? The 2.4ghz signal will definately travel that far,
causing self-interference, correct?
Travis
Microserv
Anthony Will wrote:
Answers in-line
Travis Johnson wrote:
Hi,
I'd like to go back to the specs on different radios just so I can
compare for myself...
Trango 2.4ghz:
5Mbps auto ratio
8 non-overlapping channels
10mhz spectrum per channel
-90 Receive level
15 mile range (without a grid)
External connector and dual-pol integrated antenna
$879 AP (WISP price)
$479 SU (WISP price)
Canopy 2.4ghz (regular):
7Mbps fixed ratio
3 non-overlapping channels
20mhz spectrum per channel
-86 Receive level
2.4 canopy has a -89 receive level
5 mile range (without a dish)
$902 AP (reseller price online)
$490 SU (reseller price online)
I am guessing your quoting single prices here. Now that maybe viable
for this discussion but realistically if a WISP is not financially
able to purchase in 25 packs they likely are very underfunded. So
that the information is available a 25 pack of the "Classic" 2.4 ghz
Canopy units is $6709 so if you break that down to single price that
is about $269ea + $50 for reflector for a total of $319ea.
http://www.doubleradius.com It is possible to get them cheaper then
this but you will have to deal with co-op's or ebay.com
Also I would never install a unit with a 60* pattern (Trango or
Canopy). Just include the$50 for a reflector or stinger from
http://www.wirelessbehive.com
Based on the information from Mike, I could not use Canopy. In
several areas, I have 4-5 towers located within 5 miles of each
other.... how do I do that with Canopy? With Trango, I use a
different channel for the sector pointing toward another tower
(frequency planning and coordination is very important) and
everything works great. Is there a solution for this with Canopy?
This is where GPS sync comes in. You can point two different tower
locations on the same frequency at each other and they will not
interfere with each other. This is how it is possible to do a 6 AP
cluster on one tower with only 3 non overlapping channels.
Also, by using only a 10mhz spectrum per channel, Trango's channel 1
and channel 8 are actually outside the reach of Canopy and 802.11
(for the most part) and thus can almost always be used in a noisy
environment.
Remember with Canopy you generally don't have to avoid interference.
Find the cleanest channel and 90% of the time you will be the few db
louder then the noise that you need to make a viable link.
Anthony Will
Broadband Corp
Travis
Microserv
Mike Bushard, Jr wrote:
Well, so far as we can tell the only thing that can kill canopy, IS
CANOPY.
We have put it up against WaveRider, Alvarion, and 802.11b. They
all fell of
the face of the earth.
We have 16 tower sites deployed, all 900Mhz and 2.4, over 1000 CPE
and more
on the way. (I realize there are many people bigger than us.)
We use a mix of MTI Omni's, MTI or Tiltek 120deg Sectors (MTI for
Horizontal
and Tiltek for Vertical) and integrated 60deg sectors (I really
wish someone
would come out with a descent H-pol as I don't like the integrated
antenna)
with 900. Cyclone Omni's or 120deg sectors on 2.4.
Here is what I have found with GPS Sourced Sync vs. Generate Sync:
If you want channel reuse you need GPS sourced sync.
If you have a tower more than 8 miles away, you need to use different
channels no matter what, even with GPS sourced sync you still have
speed of
light issues from tower to tower.
Can you Generate sync and deploy multiple AP's in a given area,
yes. You
just need to make sure you have Frequency separation. Does this mean I
recommend it, NO.
Also even with every site GPS Synced, you still can only put so
many AP's in
a given area be for you need to go to a different polarity. At
least we know
there will never be another 900Mhz based ISP in one of our towns.....
Also on a side note, I have never found a problem with 2.4, it is
900 that
will give you problems, it just carries so far. If the noise floor was
lower, and Canopy could run at -90 we would have coverage for a
long ways.
It seems like we can always pick up a AP at -80.
YMMV.
Mike Bushard, Jr
Wisper Wireless Solutions, LLC
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On
Behalf Of Matt Liotta
Sent: Monday, September 25, 2006 5:07 PM
To: WISPA General List
Subject: Re: [WISPA] vendor specs -- Jon
Patrick Leary wrote:
I'm speaking about multipoint matt, not ptp. The dedicated ptp you
are
doing is by far the exception. Canopy is designed, built, and sold
to be
primarily a pmp system. I've never met or heard of a Canopy pmp
network
of any scale that did not require GPS.
I'd be interested in further explanation on this topic. We have
some Canopy pmp and haven't found the lack of GPS a problem.
Granted we don't have a large amount of pmp, but I would certainly
like to understand any future pain before we experience it.
-Matt
--
WISPA Wireless List: [email protected]
Subscribe/Unsubscribe:
http://lists.wispa.org/mailman/listinfo/wireless
Archives: http://lists.wispa.org/pipermail/wireless/
