>From where to Ideal? Kirkland?
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com ----- Original Message ----- From: "Ken Hohhof" <[email protected]> To: [email protected] Sent: Wednesday, August 26, 2015 9:59:32 PM Subject: Re: [AFMUG] Spatial Diversity - helps how much? I bought a small WISP in 2005 and inherited a 10 mile 2.4 GHz PTP link that went over a hill, totally NLOS, using 802.11b (smartBridges APPO). Very consistent performance. How did it work? Magic. Unfortunately the next hop was to a private water tower that the owner decided was cheaper to demolish (and replace with booster pumps for the fire sprinklers) than to paint. From: Brian Webster Sent: Wednesday, August 26, 2015 8:48 PM To: [email protected] Subject: Re: [AFMUG] Spatial Diversity - helps how much? Trees that are not in the near field of either end of the path will cause some diffraction, if that is the case spatial diversity may help. While trees do cause attenuation of you are shooting over the top of some trees and not directly though the main mass of the trees you can sometimes get this diffraction effect somewhat like knife edge diffraction over mountains and/or sharp edges like buildings. When I worked on the EarthLink Philly project we tested this phenomenon on 5 GHz paths. Canopy 100 stuff would not make link, no near field tree obstructions but no visible line of sight but path profile showed possible but with trees. Same paths using Alvarion at the time would make link and move data. Only difference was an OFDM platform over the Canopy. We came to the conclusion that the OFDM was able to make use of the scatter diffraction over/through the tops of the trees. Thank You, Brian Webster www.wirelessmapping.com www.Broadband-Mapping.com From: Af [mailto:[email protected]] On Behalf Of Adam Moffett Sent: Wednesday, August 26, 2015 12:27 PM To: [email protected] Subject: Re: [AFMUG] Spacial Diversity - helps how much? My biggest hurdles are protruding terrain features and trees. One of my colleagues asserted that having the two spatial paths might provide two chances to find a path through the woods. It sounds simultaneously plausible and crazy, and I don't have enough background knowledge to say one way or the other. We could surely get reflections off the ground, but I've always thought of trees as a source of attenuation rather than reflection --maybe that's too simplistic, but most of the time I'd bet it's close enough to the truth for practical purposes. If spatial diversity is mostly about fighting multipath interference, then adding 6db to the link budget is not appropriate, and it sounds like it would be more fair to say that specific circumstances that might weaken your signal won't weaken your signal. Which means Telrad's "only helps a little" is the more accurate response. Is there any downside? Any circumstance where spatial diversity hurts you? I can tell you cost is not a problem. The material cost is actually lower to buy two dual pol sectors compared to the 4x4, but you have a little more labor in assembly.....cost wise it's a wash. On 8/26/2015 11:47 AM, [email protected] wrote: Historically, spatial diversity was used on long paths over non varying terrain, like deserts and lakes. Things that give off what is called specular reflections and weather refraction effects. Shooting from mountain to mountain over a bowl shaped valley is pretty bad for multipath. For short distance WiFi, I would think it may be helpful with moving reflectors like people and cars. From: Cameron Crum Sent: Wednesday, August 26, 2015 9:38 AM To: [email protected] Subject: Re: [AFMUG] Spacial Diversity - helps how much? Spatial diversity is primarily used to combat multi-path. If you have clear los, your chances of bad multi-path are fairly small and you probably won't see a lot of benefit. If you have a lot of objects between you and the tower that can cause reflections, then it will help more. Simple enough? On Wed, Aug 26, 2015 at 10:19 AM, Adam Moffett < [email protected] > wrote: There are a couple of products out there selling 4x4 MIMO (Telrad is one, but there are others). In Telrad's case, two of the chains have a time offset from the other two, so you get two chains on each of two polarities. Their default antenna is a single sector antenna with 4 N-connectors on it, so there's no significant spacial diversity. In the past it's been suggested that we use two dual pol sector antennas and space them 3 feet apart to get spacial diversity. When I asked why they do the single antenna, a source at Telrad told me that spacial diversity "only helps a little". The party selling us the two panels considers it to add 6db when they run coverage projections. I suspect any gain from spacial diversity is going to depend on a lot of circumstances and I doubt it could be as simple as adding 6db. I'm wondering if anyone here has any opinions on the topic? Maybe even facts :) (I'm sort of eyeballing a certain guy in Utah who designs antennas and isn't trying to sell me anything.)
