Re: [WISPA] Licensed Backhaul options

Mon, 20 Mar 2006 16:59:59 -0800

Title: Message
Dustin,
 
I'd like to see those screen shots.
 
In case interested...
 
I had reviewed the spec sheet of Andrews 3 ft, Gabriel 2ft, 2ftHQ, 4ft, 4ft HQ and compared.
Understanding that this is theory based on spec sheets being accurate, this was my finding...
 
2ft HQ dish did great (best of 3) to prevent colocation interference.
However, 4ft dish (regular) did much better at preventing front interference.
However, Andrews 3ft, came pretty darn close to the performance of 4ft dish, at 3ft ease, price, and windload.
Of course 4ft HQ dish beat them all in all categories, but at a hefty price and windload.
 
Analyzing three sites in question, each site was better off with a different one of the antennas based on the need of the site.
Site 1: The 3ft, needed high gain, some noise in front, but no other antennas on the roof.
Site 2: The 2ft HQ, lots of colocation noise, but landlord not allow 4ft dish. (used 2ft over 3 ft, as msot noise was colocation based)
Site 3:  The 4ft HQ, lots of colocation noise, 4ft mounting OK, lots noise in front, and every db counts.
 
I'm going to put them to the test next week when they arrive.
 
Here were actual compulations from spec sheets, (averaged data).
 
Degrees are number of degrees off center, not total beamwidth.
 

2ft compared to 2ft HQ
 
aboutsame until...
-60 deg off center,4 db more for HQ
-90 deg off center, 12 db more for HQ.
-180 deg off center, 9 db more for HQ
 
summary: no benefit to combat front end noise. Great for colocations noise, and limited antenna seperation.
 
4 ft compared to 2 ft HQ
 
4 ft much better than 2ftHQ till 58 deg. 
 
at 7 deg,  4ft 9 deg better
at 40 deg, 4ft 5 deg better
 
at 58 deg both antennas are equal at -32 isolation.
 
at 90 deg, 2ftHQ -40 db isolation, 8 db better than 4ft
at 180 deg, 2ftHQ -46db isolation, 4 db better than 4ft
 
Summary: 4 ft better for front end interference.
however, 2ftHQ still significantly better for colocation noise.
 
Andrews 3 ft - 4ft
 
at 7 deg, 4ft 9 db better than Andrews
at 10 deg, 4 ft 5 db better than andrews
at 20 deg, 4ft 8 db better than andrews
 
at 40 deg, both about the same at -32db.
(steady slope to 90)
at 90 deg, andrews -38, 6 db better than 4 ft.
 
at 119-180 deg, both at about -42.
 
Summary, 4 ft better for front interference (below 35 degrees).
however, Andrews has equal colocation protection, and better verticle speration, and 4ft dbi, at 3ft ease and windload.
  
Andrews 3 ft - 2ft HQ
 
at 4 degrees, andrews -13, 11 db better than 2ftHQ
at 7 deg, about the same
 
at 10 deg, andrews 5 db better
at 12 deg, the same
at 20 deg, andrews 4 db better
at 40 deg, andrews 3 db better
at 90 deg, 2fthq (-40) 2 db better
at 180 deg, 2fthq (-46) 4 db better
 
Andrews 3ft - 4ftHQ
 
at 3 deg, 4fthq 3 db better
at 7 deg, 4fthq 8db better
at 20 deg, 4fthq 4 db better
at 40 deg, 4fthq 5 db better
at 72 deg, 4fthq 7 db better
at 95 deg, 4fthq 15 db better
 
Note: It is understood noise will rise with antenna gain, so increased dbi should not be considered regarding interference resilience, for side of link with interfrence.
However, increased dbi will still add benefits on other side of link.
So a higher gain antenna would still be more beneficial in some respects.
So a smaller HQ antenna may not help the other side as well as the higher gain antenna.
 
 
Tom DeReggi
RapidDSL & Wireless, Inc
IntAirNet- Fixed Wireless Broadband
 
 
----- Original Message -----
Sent: Monday, March 20, 2006 3:09 PM
Subject: RE: [WISPA] Licensed Backhaul options

Actually using both poles increases the sensitivity of the radio at least by the way the Orthogon/Canopy 300’s operate when in single payload. 

 

The dual payload option decouples this feature and you really have two radio’s sending information on the same dish, one in H and one in V.  If you look at the release notes any time the radio’s are in dual payload you need more receive sensitivity then in single payload.  (makes sense right?) Only the Spectra/Canopy300’s operate in dual payload mode.  

 

The Gemini/Moto 60’s operate in single payload all the time giving them a very high system gain by allowing them to combine both poles into a single signal for processing. 

 

So to answer your question, Yes dual pole can combat noise by increasing system gain really just creating more C/I.   A while ago I told you that I was testing some of the new Gabriel High Performance dishes.  When I get some time I have some screen shots that I’ll post that I think you’ll find very interesting. 

 

Dustin

 

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tom DeReggi
Sent: Monday, March 20, 2006 2:52 PM
To: WISPA General List
Subject: Re: [WISPA] Licensed Backhaul options

 

Dustin,

 

You have showed how using both Pols, can reduce the sensitivity requirements of the radio, able to extend link distance, and likely aid in NLOS (that could degrade signal) as well.

However, when Tackling noise, that doesn't really help, does it? As the sensitivity drops, that RSSI level also gets closer to the noise floor, in many cases cancelling out the benefit.

 

So my question to you is.... have you seen the Dual Pol config help combat the noise? Meaning... Allowing the radio to operate closer to the noise floor at high modulations. Maybe by rebuilding the wave by comparing them? For example, Proxim's circular pol solution, often allows it to operate closer to the noise floor because if it.

 

Tom DeReggi
RapidDSL & Wireless, Inc
IntAirNet- Fixed Wireless Broadband

 

 

----- Original Message -----

Sent: Sunday, March 19, 2006 6:47 PM

Subject: RE: [WISPA] Licensed Backhaul options

 

Charles you make a good point, but I’m going to throw a “but” in here:

 

 but the Orthogon / Canopy 300 radio’s will run also run at:

 

64 QAM .92 dual          -62 receive sensitivity              +18 output     (252.9 throughput)

64 QAM .75 dual          -68 receive sensitivity              +18 output    (206.7 throughput)

16 QAM .87 dual          -71 receive sensitivity              +20 output    (160.8 throughput)  

 

Full list can be found in the release notes and if you do the math on those modulations you can get some very good performance.  I do agree with you that the licensed links would make more sense, but hanging 4 foot dishes on towers becomes a very expensive task or if you have to do a non-penetrating roof mount skid, the cost difference between the sleds is big.  So we have to take in more than the cost of the radio’s, licenses, leases and dishes but put together the total cost because if you are hanging BIG dishes you’re going to dig deeper into your pocket.

 

Sorry I missed you at the show, I’m on baby watch so when she saw wireless beer and gear she called foul and took away my kitchen pass L.

 

Dustin Jurman

 

 

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Charles Wu
Sent: Saturday, March 18, 2006 5:26 PM
To: 'WISPA General List'
Subject: RE: [WISPA] Licensed Backhaul options

 

>The Spectra would be around $20k with external antennas. A licensed product is going to be at least that, and probably $5k more.

Sit back and actually think for a second about this comparison, and you'll realize that a similarly performing "unlicensed" solution will cost MUCH MORE (and be much riskier) relative to the licensed solution

 

The main difference is that the spectra requires 30 Mhz of ABSOLUTELY CLEAN SPECTRUM in both the vertical and horizontal polarities (150 Mb "Air Rate" transmits on V-pol & 150 Mb "Air Rate" transmits on H-pol -- cut off 1 polarity, you halve throughput)

 

In addition, the Rx sensitivity of the Spectra at the 300 Mb data rate (256 QAM) is -59 dB with an output power of +18 (so you'll need HUGE dishes to guarantee the link budget)

 

So, lets do a "theoretical" path calc / comparison (15 miles)

 

11 Ghz Licensed Link (100 Mb Full Duplex)

Rx Sensitivity: -76 dBm

Tx Power: +21 dBm

4' Dish: +39 dBi

 

Expected RSSI: -42.9 (>30 dB of fade margin = ROCK SOLID LINK =)

 

5 GHz Spectra

Rx Sensitivity: -59 dB

Tx Power: +18

6' Dish: +34 dBi

 

Expected RSSI: -49.4 (~10 dB of fade margin w/ 2' more of each dish)

 

Then there's all sort of "real-world" performance issues that occur with higher-order modulation schemes and license-exempt operation

 

-Charles

 

-------------------------------------------
WiNOG Austin, TX
March 13-15, 2006
http://www.winog.com

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Travis Johnson
Sent: Friday, March 17, 2006 10:03 PM
To: WISPA General List
Subject: Re: [WISPA] Licensed Backhaul options


Travis
Microserv

Charles Wu wrote: 

You don't need licensed to high throughput backhaul. For example, 
Orthogon's Spectra provides 300Mbps aggregate at a price point generally 
Less than 45Mbps licensed.
    
 
Hi Matt,
 
I am curious to see where / what you got those numbers for the Orthogon
Spectra?
 
-Charles
 
-------------------------------------------
WiNOG Austin, TX
March 13-15, 2006
http://www.winog.com 
 
 
 
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On
Behalf Of Matt Liotta
Sent: Thursday, March 16, 2006 1:28 PM
To: WISPA General List
Subject: Re: [WISPA] Licensed Backhaul options
 
 
 
-Matt
 
Bobby Burrow wrote:
 
  
I'm looking at moving to a licensed solution to increase throughput 
across one of out backhaul links that spans 5 hops. Distances between 
hops range anywhere from 7 to 19 miles.
 
We are currently using the dual nstreme Mikrotik solution and it is 
working very well, however the WRAP/RB532 solutions are only yielding 
~25Mb per hop.
 
Can anyone recommend a licensed radio manufacturer that should net us 
50Mb-100Mb per hop?
 
Thanks,
 
Bobby Burrow
East Texas Rural Net
www.etxrn.com
 
 
 
 
    
 
 

--
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/
-- 
WISPA Wireless List: wireless@wispa.org

Subscribe/Unsubscribe:
http://lists.wispa.org/mailman/listinfo/wireless

Archives: http://lists.wispa.org/pipermail/wireless/

Reply via email to