With that tree scenario, do you still feel youre protected from interference in unlicensed spectrum, or if others start using the band, will you be screwed with those signal levels?
In a situation where the customer is buried in trees, I figure the trees will probably block most interference, but upstream may be vulnerable. If its a forest town as you describe, are you counting on CBRS and the spectrum database to protect against interference? From: Af [mailto:af-boun...@afmug.com] On Behalf Of George Skorup Sent: Thursday, October 13, 2016 12:34 AM To: af@afmug.com Subject: Re: [AFMUG] i need an adult Maybe one tree up close, say 50-100 feet, is, as you describe, probably not much of an issue. Sure, you can call this NLOS. A cluster of trees a 1/4, 1/2 or a mile away, maybe you can't really see the tower, I still consider enough LOS for RF wavelength. We do these kinds of shots with PMP450 today and it works fine. This is what I'd call good nuf LOS aka near-LOS. As Ken says, sometimes this can actually improve the signal. Clip that fresnel zone just right and you can get a boost. And we did find that -/+45 does help a bit for these kinds of shots. The NLOS I'm talking about is an entire town full of trees. All up in your face. Even surrounding the tower. A forest town. Where 900 barely works. A place where we're considering doing fiber instead (preferably in the ground so we don't get that other kind of "tree fade"). That's what we're dealing with. It is nuckin futs. The power levels we're seeing really does not surprise me. Leaves are coming down too. Should've had it up a month ago. BTW, we're doing 10MHz channel width. There's too much other gear around to try to get away with 20MHz. On 10/12/2016 9:59 PM, Craig Schmaderer wrote: Ive been running baicells since june. We have over 16 links around .5 to 1 miles going through trees like they don't exist getting 80mb down. �A mile of trees might be pushing it ;) Craig Schmaderer Cell 402-380-1245 Skywave Wireless, Inc. _____ From: Af <mailto:af-boun...@afmug.com> <af-boun...@afmug.com> on behalf of George Skorup <mailto:geo...@cbcast.com> <geo...@cbcast.com> Sent: Wednesday, October 12, 2016 9:17:37 PM To: af@afmug.com <mailto:af@afmug.com> Subject: Re: [AFMUG] i need an adult � Aggregate. The guys were out with one CPE moving around. The -115 RSRP location was giving about 140ms latency. From what I heard anyway. Doesn't surprise me seeing as how the uplink bandwidth was pretty much nothing. We're going to try switching a couple customers from 900 FSK over to this thing tomorrow. On 10/12/2016 9:09 PM, Ken Hohhof wrote: 7 Mbps per sub or aggregate?� I assume aggregate.� Well, OK, if you only have 1 sub they�re� the same. � Yes definitely, log CINR and modulation. � Also might want to do a ping test and measure min/avg/max latency and pkt loss. � � From: Af [mailto:af-boun...@afmug.com] On Behalf Of George Skorup Sent: Wednesday, October 12, 2016 8:09 PM To: af@afmug.com <mailto:af@afmug.com> Subject: Re: [AFMUG] i need an adult � So far we're seeing no better than about -88 RSRP. Most places we tested was right around -100. All Non-LOS. Very Non-LOS. Within a mile. And the nitwits didn't get my memo that I wanted to collect CINRs and mod rates too. They were able to still get about 7Mbps down (less than 50kbps up tho) with about a -115 RSRP. And I did double check that we're not exceeding the 1W/MHz EIRP cap. We're going to do some more testing tomorrow. On 10/11/2016 8:58 AM, Ken Hohhof wrote: I hope it works for you through the solid mile of trees, I�m skeptical though.� It seems to me like a golfer trying every new brand of club and ball that comes out hoping to find one that goes straight through a mile of trees to the green.� You may get incremental improvement, but I know people in that situation whose cellphones don�t work. � � From: Af [mailto:af-boun...@afmug.com] On Behalf Of George Skorup Sent: Monday, October 10, 2016 11:37 PM To: af@afmug.com <mailto:af@afmug.com> Subject: Re: [AFMUG] i need an adult � I haven't changed or turned on anything new facing you. And I don't do alternate freqs on the 450's even though sometimes LBT goes to hell. Things have really calmed down since moving the entire 450 network from 13.2.1 to 14.1.2 though. We are going to be testing a Baicell tomorrow in an isolated area and not even facing you. Just FYI. I have it down in the lower half of the band to step on the UBNT shit which I don't really care about anyway. A straight mile deep trees, so I doubt we'll even see it on any other sectors. On 10/10/2016 11:11 PM, That One Guy /sarcasm wrote: Ive been beating the hell out of my network this last week with the 320 going to shit. The UBNT 3ghz hasnt gotten any worse that it was I went as far as shutting down the majority of the 3 ghz to get spectrums. Went through and coordinated the network based on those. i added QOS profiles with harq, it got me management of most of the CPEs, the customer experience is deplorable, theyre all probably trump supporters. Im going to let this sit for a few more days to see if Im just fighting the atmosphere. If it doesnt resolve, Ill need to hail mary and expert for a paid gig to get on my network with me for a few hours one evening soon and put into words for my boss just why it is he should not hire incompetent boobs like me to run his network, and hopefully offer some advice on how i can save face. im getting in a bunch of packetflux so all the 320 is synced (there are a few outliers that never got sync as other projects took priority and I have a couple base units that lost communication with the syncinjectors. I figure if I get everything on the same sync product I unsure theres not some diode between manufacturer products offsetting the puls by a few nanoseconds. I see a forklift in the near term of ubnt 3ghz to epmp 5ghz to free up spectrum for 450, which i assume will destroy the remaining 320. � � Please hit me offlist if you will be available for a contract � -- If you only see yourself as part of the team but you�don't�see your team as part of yourself you have�already failed as part of the team. � �
