Shooting a 900mhz across a lake?

[ducklight]

I am trying to shoot a 900mhz signal across a lake from a water tower to multiple points. I am using

http://www.trangobroadband.com

equipment, specifcally the M900S radio. The water tower is approx 150ft above the normal water level. The other endpoints are fixed location and vary in heights from 10' to 200' above the normal water level. The lake level varies 10-15ft. A distance of 1/2 mile to 2 miles is typical. I was using the builtin antenna on the 900s but have tried a 13db yagi. Due to recent flooding the lake level has raised significantly and I have lost reception. I am getting approx -73db gain but I can't hold a baseid. I think the problem is from multipath reflection from the water but I am not sure of the best solution to solve my imediate problem and a longterm solution. Any ideas?

 

[zeitghost]

There was a similar question to this some months ago... it'll be further down in this forum (maybe)... his problem was a river.

 

[pstuckey]

Circular polarization offers some protection against multipath fading. When a circularly polarized signal reflects off a surface the sense reverses (RHCP becomes LHCP and vice versa).

For instance if both Rx and Tx antennas are RHCP, the direct path is RHCP and is received normally, the reflected path (single reflection) is LHCP and decreased by the cross polarization rejection of the Rx antenna (typically ~20 dB).

Peter

 

[Comcokid]

Zeitghost - your memory is correct.
Ducklight - see thread239-183427

Are you loosing it from the water tower (150') to one other point, or to all of the other points? If you were loosing it to a point that was 10' above the water, multipath due to the lake might be one of a number of suspects, but to many of your other points, especially the one 200' above the lake, indicates some other kind of problem.

Has the foilage from trees expanded to cover the line-of-site? Additional moisture caused alge to grow over the enclosures (blocking the internal antennas)? Moisture building up in the enclosures (creating a condensed moisture over the inside of the enclosure). Do the links work at one time of the day and not another? Has a local new cellular tower gone-up since you installed the system?

 

[VEBill]

The wavelength of 900 MHz is roughly one foot. You can do the geometrical analysis, but multipath would have come-and-gone many times if the lake surface was going up and down 10-15 feet. Also, it would affect different endpoints differently (very unlikely that all the delta path lengths would all vary in phase). Also, everything involved tends to vary sufficiently that multipath normally isn't a hard failure (it comes and goes).

So, the problem is probably something else.

 

[biff44]

This may not be an antenna problem but more of a system problem. Multipath, in itself, is not a huge problem. It tends to either constructively or destructively interfere with the signal at the receiver. If your receive has sufficient gain, it will simply adjust the receiver gain to continue to receive the signal.

However, if you are sending a wide bandwidth signal (lots of Mbps), then at one frequency in the transmit spectrum you might be seeing a different path loss and path delay that at another frequency in the transmit frequency spectrum. A simple AGC gain control in the receiver can not fix this type of "dispersive fade". You would want to be using a receiver that has an "adaptive equalizer" to correct for gain and phase variations across the transmit channel. Perhap you need to change the transmit and receive boxes to be more modern types.

 

[bmnet]

If you're dealing with multiple receivers, try moving one of them and hopefully you'll move it away from the 'null'. Most likely the tx antenna's pointing in a different direction to cause acute reflection angle (the change in water level too could necessitate an adjustment of the tx antenna)

 

[jediknight7]

You might try 900MHz solution from Ubiquiti Networks (

http://www.ubnt.com).

Their stuff works unbelievably well.