900 MHz waverider & ISP problems

Hello All,

I am currently using a waverider EUM3000 and running into issues where I lose my signal on and off. I am having problems pin-pointing the items that are causing interferences.

When aligning my antenna I can achieve about 77+/- dbm but can never get it to stick/stay constantly. It will be working good but then numerous times throughout the day it drops in and out.

I do have a lot of trees by me but I have put together a tower that should be putting me above the tree line.

I was under the impression that it had to be something on the ISP side but they claim not to be causing any problems and I am just not sure...so I thought that I might ask around about things that could be causing my problems.

Any information is greatly appreciated. Thanks

Reply to
nurp
Loading thread data ...

How about some numbers? Range to access point? Type of antennas in use? Length of coax run (if any)? Approximate path profile (how bad is the line of sight ignoring the trees)? Extra credit... Exact lat-long and antenna height of both ends so I can calculate the fade margin. |

formatting link
wouldn't expect you to actually peform this test, but the calcs can be done with sufficient information.

What makes you think it's interference and not multipath? I think (not sure) that -77dBm is a fair signal stength. However, if it's varying all over the place, my guess(tm) is that you have either more than one path between your EUM and the CCU, or that the trees in between are moving in the wind. I don't have any experience with Waverider, but I sure have had some with 900MHz in the forests with Metricom. Getting the signal to stay put was a big problem as the path through the forest was never direct. 900MHz would go though the branches but not the trunks. Put enough trunks in the way and nothing would work. Tree trunks swaying in the wind would open vertical "slots", which would dramatically improve propagation for a few seconds, and then die as the trunk swayed the other way. The only solution was to get as much gain as possible at the client side antenna and hope for the best. Moving the antenna around was also useful, but was by no means a guaranteed solution.

How many trees in a "lot"? How many miles of trees are you trying to penetrate? What type of tree? Thick forest or thin? Ignoring the trees, do you have line of sight or is there a hill, mountain, building, or something in the way?

Good plan. I like to do the numbers first, before doing the field trip. Calculate the line of sight and the fade margin first.

Take your EUM and go for a drive. Get a 117VAC inverter for your vehicle, a laptop, perhaps a spare antenna, and take it to some place with a clear line of sight to the CCU at approximately your unspecified range. Duz it work? Is it stable? If not, there's something broken in your EUM. It's probably not the CCU or there would be a rioting mob of irate customers at the ISP's door.

If that works, put some trees in the way, but no mountains or hills. Just trees. Duz it still work? Monitor your signal strength. Is it stable? Keep going until it craps out. Is it about the same as your location or farther?

Reply to
Jeff Liebermann

Thanks for you reply.

The base station is probably no more than 100' and I am around 60'. I am using a yagi antenna, not sure about the gain (maybe 14 db)...the cable is between 50' or 60'. The tallest tree I have by me is probably around 60' & I should be well under 3 miles away. There are no buildings, hill or mountians in the path just a large amount of trees, maybe about the size of several city blocks or so.

I might already have the lat/long and I guess I can take a drive this weekend.

Thanks for your help.

Reply to
nurp

I guess that's the altitude. Could I trouble you for the distance?

How many elements and I'll calculate the gain.

Ouch. Might be very lossy that long. What type of cable? If unavailable, measure the diameter.

That's not too horrible. However, I have no idea how big is a city block. I live in a rural area. Can we do this in calculateable units of measure, please?

Don't forget that wireless requires more than an optical line of sight. The Fresnel zone is considerably wider at midpoint at 900MHz than optical. That can be calculated once you supply the missing distance. If the Fresnel zone hits the ground, you're in trouble.

With the exact lat-long, I can use a program called "Radio-Mobile" to plot the path profile. Your Waverider dealer should also be capeable of calculating the fade margin and doing a path profile.

Reply to
Jeff Liebermann

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.