How many people use 900APs with an omni antenna? What kind of performance can be expected? Does the omni allow too much noise, and unstable links?
If you are thinking about deploying a 900MHz omni, I would first do a spectrum analysis of the area you are looking at. This will give you a good idea as to what else is operating in the band.
Keep in mind that pretty much everyone else using 900 gear is going to be running vertical polarization. This group covers anyone from a 900MHz cordless phone, water treatment plants, power companies, farms, and high-power paging transmitters. My experience has been that most of these are FHSS or some type of hopper, with the exception of the pager transmitters, as they just dish out 500 or more watts of sterilizing goodness. :lol:
You will usually find that everyone that needs to do any WISP type operations will run horizontal polarization, so that they can get the xpol seperation.
Most of the other users that I listed will be transmitting data in short bursts. An example is a water treatment facility needing to send and recieve telemetry data from one of the lift stations. The lift station is getting full, so it sends a short blip of data to the master site and a computer there knows that it needs to start a pump or send an alarm. The master site might send a blip back telling the station that it needs to go ahead and activate the pump.
This whole scenario takes place in a matter of seconds, and it is not a continuous stream. As such, if you are running VPOL and in the same band, chances are you are going to experience interference and it might manifest itself as re-reg’s. As such, spend some serious time running your SA test.
900MHz usually is used because it is easy to get components and they usually are cheap. The other alternative to the telemetry units running 900MHz is VHF telemetry radios. These wont cause problems for you, beacuse they are in a totally different frequency.
We run RFS Celwave omnis on our 900APC’s, some with down tilt some with out with good luck. We have 3 customers 20-25 miles out on various sites that connect solid, one at -64ish @ 22 miles out. Eventually, we will be changing our config from one AP/omni per site to 3 APs and 120 degree panels. This will allow us to work around interference easier but you gotta start off somewhere.
mhussey wrote: We run RFS Celwave omnis on our 900APC's, some with down tilt some with out with good luck. We have 3 customers 20-25 miles out on various sites that connect solid, one at -64ish @ 22 miles out. Eventually, we will be changing our config from one AP/omni per site to 3 APs and 120 degree panels. This will allow us to work around interference easier but you gotta start off somewhere.
I dug all through the RFS site. Do they make a horizontal polarized 900 omni? If yes what part num?
I don’t think they do.
You know what’d be funny is seeing a Penetrator at the top of a tower with a Canopy radio connected to it…haha
Matt, Im not sure I would have to look. We are running VPol.
yes, the last mile gear omni / cyclone 900 mhz omni is h-pol.
we have two of them active - 12 subs on one , 18 on the other. our 120 panels have 24 and 19 respectively…the sectors allow you to control your interference a lot better…but the omni is good for a rural area where you’re sure you have no interference. as they’ve all said, do a spectrum first using the omni - it’ll probably work well until your noise floor rises or the number of subs you have on your 900 mhz ap rises…then you’ll need to sectorize.
we started our company with 2 900 mhz omnis…we now have an omni on one tower, 2 120 sectors on another, a 120 sector on another, a 120 sector on another, and an omni on another (and we’re having trouble in that area…we probably need to sectorize soon).
here are the stats from the 900 omni we have in a rural area:
it hasn’t been restarted in 9 days and as you can see, we have some work to do…
LUID: 002 : State: IN SESSION (Encrypt Active)
Site Name : Drake
Software Version : CANOPY 7.3.6 Oct 24 2005 12:06:56
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 54 (approximately 1.50 miles (7938 feet))
Session Count: 11, Reg Count 11, Re-Reg Count 5
RSSI (Avg/Last): 1263/1263 Jitter (Avg/Last): 5/5 Power Level (Avg/Last): -79/-79
DnRate(A): 20000 DnLimit(A): 500000 UpRate(A): 64 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 18 Rate 2X/1X
LUID: 003 : State: IN SESSION (Encrypt Active)
Site Name : Hinson
Software Version : CANOPY 7.3.6 Oct 24 2005 12:06:56
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 27 (approximately 0.75 miles (3969 feet))
Session Count: 3, Reg Count 2, Re-Reg Count 5
RSSI (Avg/Last): 1259/1277 Jitter (Avg/Last): 4/4 Power Level (Avg/Last): -80/-79
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 19 Rate 2X/1X
LUID: 004 : State: IN SESSION (Encrypt Active)
Site Name : Baptist Church
Software Version : CANOPY 7.3.6 Oct 24 2005 12:06:56
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 82 (approximately 2.28 miles (12054 feet))
Session Count: 3, Reg Count 1, Re-Reg Count 2
RSSI (Avg/Last): 1314/1312 Jitter (Avg/Last): 3/2 Power Level (Avg/Last): -78/-78
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 20 Rate 2X/2X
LUID: 005 : State: IN SESSION (Encrypt Active)
Site Name : Goodwin
Software Version : CANOPY 7.3.6 Oct 24 2005 12:06:56
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 68 (approximately 1.89 miles (9996 feet))
Session Count: 6, Reg Count 10, Re-Reg Count 5
RSSI (Avg/Last): 1255/1262 Jitter (Avg/Last): 4/4 Power Level (Avg/Last): -80/-80
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 21 Rate 2X/1X
LUID: 006 : State: REGISTERING
Session Timeout: 2, AirDelay 149 (approximately 4.15 miles (21903 feet))
Session Count: 664, Reg Count 807, Re-Reg Count 757
RSSI (Avg/Last): 1068/1000 Jitter (Avg/Last): 5/5 Power Level (Avg/Last): -84/-86
DnRate(D): 6978 DnLimit(D): 500000 UpRate(D): 22 UpLimit(D): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 22 Rate 1X/1X
LUID: 007 : State: IN SESSION (Encrypt Active)
Site Name : Graham
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 29 (approximately 0.81 miles (4263 feet))
Session Count: 1, Reg Count 1, Re-Reg Count 0
RSSI (Avg/Last): 1691/1669 Jitter (Avg/Last): 2/3 Power Level (Avg/Last): -59/-60
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 23 Rate 2X/2X
LUID: 008 : State: IN SESSION (Encrypt Active)
Site Name : Blake
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 33 (approximately 0.92 miles (4851 feet))
Session Count: 1, Reg Count 1, Re-Reg Count 0
RSSI (Avg/Last): 1413/1407 Jitter (Avg/Last): 2/2 Power Level (Avg/Last): -71/-72
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 24 Rate 2X/2X
LUID: 009 : State: IN SESSION (Encrypt Active)
Site Name : Johnson
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 277 (approximately 7.71 miles (40719 feet))
Session Count: 422, Reg Count 513, Re-Reg Count 479
RSSI (Avg/Last): 1112/1093 Jitter (Avg/Last): 5/5 Power Level (Avg/Last): -83/-84
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 25 Rate 2X/1X
LUID: 010 : State: IN SESSION (Encrypt Active)
Site Name : Skinner
Software Version : CANOPY 7.3.6 Oct 24 2005 12:06:56
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 112 (approximately 3.12 miles (16464 feet))
Session Count: 2, Reg Count 2, Re-Reg Count 0
RSSI (Avg/Last): 1246/1249 Jitter (Avg/Last): 4/3 Power Level (Avg/Last): -80/-80
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 26 Rate 2X/1X
LUID: 011 : State: IDLE
Session Timeout: 0, AirDelay 131 (approximately 3.65 miles (19257 feet))
Session Count: 316, Reg Count 929, Re-Reg Count 1773
RSSI (Avg/Last): 1005/1033 Jitter (Avg/Last): 10/5 Power Level (Avg/Last): -86/-85
DnRate(D): 6978 DnLimit(D): 500000 UpRate(D): 22 UpLimit(D): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 27 Rate 1X/1X
LUID: 012 : State: IN SESSION (Encrypt Active)
Site Name : Mount
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 24 (approximately 0.67 miles (3528 feet))
Session Count: 2, Reg Count 1, Re-Reg Count 1
RSSI (Avg/Last): 1612/1621 Jitter (Avg/Last): 3/3 Power Level (Avg/Last): -61/-61
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 28 Rate 2X/1X
LUID: 013 : State: IN SESSION (Encrypt Active)
Site Name : Mason
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 156 (approximately 4.34 miles (22932 feet))
Session Count: 1, Reg Count 1, Re-Reg Count 0
RSSI (Avg/Last): 1326/1319 Jitter (Avg/Last): 4/5 Power Level (Avg/Last): -77/-77
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 29 Rate 2X/2X
LUID: 014 : State: IN SESSION (Encrypt Active)
Site Name : McCravy
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 152 (approximately 4.23 miles (22344 feet))
Session Count: 96, Reg Count 55, Re-Reg Count 105
RSSI (Avg/Last): 1302/1297 Jitter (Avg/Last): 3/3 Power Level (Avg/Last): -78/-78
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 30 Rate 2X/1X
LUID: 015 : State: IN SESSION (Encrypt Active)
Site Name : Hackney
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 122 (approximately 3.40 miles (17934 feet))
Session Count: 18, Reg Count 21, Re-Reg Count 14
RSSI (Avg/Last): 1182/1184 Jitter (Avg/Last): 5/5 Power Level (Avg/Last): -82/-82
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 31 Rate 2X/1X
LUID: 016 : State: IN SESSION (Encrypt Active)
Site Name : Reaves
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 43 (approximately 1.20 miles (6321 feet))
Session Count: 3, Reg Count 4, Re-Reg Count 0
RSSI (Avg/Last): 1305/1313 Jitter (Avg/Last): 4/3 Power Level (Avg/Last): -78/-78
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 32 Rate 2X/2X
LUID: 017 : State: IN SESSION (Encrypt Active)
Site Name : Woods
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 156 (approximately 4.34 miles (22932 feet))
Session Count: 8, Reg Count 7, Re-Reg Count 6
RSSI (Avg/Last): 1222/1213 Jitter (Avg/Last): 5/6 Power Level (Avg/Last): -81/-81
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 33 Rate 2X/1X
LUID: 018 : State: IN SESSION (Encrypt Active)
Site Name : Jay Phillips
Software Version : CANOPY 7.2.9 Jul 23 2005 01:49:03
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 263 (approximately 7.32 miles (38661 feet))
Session Count: 41, Reg Count 54, Re-Reg Count 35
RSSI (Avg/Last): 1196/1196 Jitter (Avg/Last): 6/4 Power Level (Avg/Last): -81/-81
DnRate(A): 6978 DnLimit(A): 500000 UpRate(A): 22 UpLimit(A): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 34 Rate 2X/1X
LUID: 019 : MAC: 0a-00-3e-92-1e-95 State: IDLE
Session Timeout: 0, AirDelay 370 (approximately 10.30 miles (54390 feet))
Session Count: 10, Reg Count 7, Re-Reg Count 3
RSSI (Avg/Last): 1247/1249 Jitter (Avg/Last): 4/6 Power Level (Avg/Last): -80/-80
DnRate(D): 6978 DnLimit(D): 500000 UpRate(D): 22 UpLimit(D): 100000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 35 Rate 1X/1X (he powers off when not using)
WOW
What is your noise floor like there??? Those subs running close to -81 or more would never work here.
Are you using the 120* MTi sectors?
we loose the subs when they drop to about -83 or -84. I’ve seen some survive as low as -87
this is a 900 mhz cyclone omni. (at this tower)
unless you’re in a truly rural area, sectorize!
oh, and don’t use v-pol. vertical polarized 900mhz ISM traffic is pretty bad,
even in rural areas. utility companies use v-pol ISM-band telemetry send data from water towers, remote power sites, etc.
our choice to use horizontal polarization has proven itself.
as far as sectorizing - if you can afford it, do it. even if you can’t - do it.
we had one 900mhz omni site, but after 40 or so customers, it’s just bad news. always remember, omnidirectional antennas are simply a bad idea. at least if you’re sectorized, you can change the channel on a single sector to dodge new interference, while an omni - well, you’ve got a whole new world of headache when a new interference source comes into town.
we’ve got several hundred 900MHz subscribers, all sectorized - we’ve got no omnis anywhere anymore, and that’s been the best decision we’ve made in our 900mhz deployments.
some sites use 120 degree sectors, while some use 60 degree panels.
bottom line, sectorize and h-pol.
Are you using H-Pol Panels and then flipping yagis/antennas to H-pol? Just curious as to what your setup is. I plan on going away from omnis as soon as possible and run 3 120 degree panels and planned on going V-pol. If not I would have to revisit each customer to rotate antennas 90 degrees. Has anyone done a comparison test on coverage between a V-pol vs. H-pol setup?? I realize the beamwidth maybe narrower on H-pol and possibly decrease signal levels but at the same time it would increase isolation from interferenc.
Depends on your area. As mentioned, 90% of other 900MHz signals are vertical.
If it were me, I would go Horizontal and go through the pain of converting all the customers. Going vertical is just placing a limitation on the system. As expensive as this stuff it I want the maximum possible performance from every AP.
Agreed. Going VPOL is nothing but trouble in most areas. The only sure way to really know if it is going to work is to study the area.
Mhussey,
So you are planning 3 120* VPOL sectors?
I take it you have done a good examination using a spectrum analyzer or service monitor?
Where do the Interference mitigation radios come into play vs the standard radios? Do the Motorola AP’s & SM’s use Horizontal or Vertical polarization?
900 integrated AP and SM units are HPOL. Connectorized versions allow you to choose VPOL or HPOL.
Everything I use is HPol.
I have tried 900 Omnis from Pac Wireless, Antel Inc (the brand Cyclone uses), and MTI. So far our best performance has been the MTI antennas. The Pac Wireless and Antel/Cyclone seem to perform about the same. I had a MTI on a tower and replaced it for an Antel for more vertical beam width and almost everyone lost a good 6-7 dbm when that happened. A few subs who were well outside the beam width of the MTI improved a bit, but overall there was a noticeable signal drop.
My most used omni has over 50 subs on it and the signal levels range from -48 to -81. (Noise floor at -80 for 902-922) No one has worse than 99% downlink and uplink effeciency on a 10 sec link test.
I think omnis are great for areas where you do not need a large vertical plane due to elevation differences or where there just are not that many subs in the area. My best shot yet is 11 miles from an MTI omni to a 17 Yagi at -70ish dbm.
Some of my other antenna experience:
2x superpass 120* up which do not work well but they get by for a max range of 3 miles. These seem to have random null spots.
4x Integrated APs covering 360* quite well from the top of a 140’ tree.
1x Integrated AP covering about 80* up to 6 miles out. I also have a few people off the back side of this up to 1 mile away with only a few reregs in 50+ days.
1x 180 MTI which works wonders off a guys chimney, quite a few people off the edges of this at 7 miles getting -65dbm. This used to be a Superpass 180* which had all sorts of problems and much worse signal strength.
I also have 4x 90* MTI antennas which I have spent some time playing with but not in production. These work really nicely too. They are planned to go on a commercial tower on a prominent hill.
At this point my preference is MTI for antennas unless I am using Integrated APs in trees or for small areas.
Before you put up 3x 120* you might want to make sure you have all 3 channels available. I cannot use anything above 922 in my area. A spec analysis when the unit is pointing at the ground shows -40 or worse. So I have been forced to work with omnis or 4x 90.
Also I would not recommend the use of a 180 to anyone. You might as well put up an omni or 2x Integrated APs. According to the antenna patterns you end up with the less power at 180 degrees than an omni or the same with an integrated, and from actual testing it seems an integrated at 90* actually performs better, plus the integrated is higher gain in the center and much smaller profile.
I’ve had good success with Pacwireless. They make a HPOL omni…