Sure, circular polarization ought to be great in this application, but as you pointed out, circularly-polarized antennas aren't readily available (or as cheap).
(btw, Theodore Rappaport published an interesting paper a few years back that showed that the multipath for indoor networks was reduced significantly with circularly-polarized antennas. Apparently, paths with an odd number of bounces had the polarization reversed. Paths with an even number of bounces did not, but since the loss from two or more bounces is greater than from one bounce, the multipath energy arriving at the receiver was greatly reduced!) Yes, I should have mentioned that I didn't bother to check the path loss on a path of this length, and I haven't bothered to check that there's sufficient clearance along the path for the first Fresnel zone. Andy should check that if he didn't do that already as part of the link design. Other things to look out for on over-water paths are ducting due to temperature inversions over the water (which can be avoided if the path is made to have an angle of 2 degrees or more to the inversion layer), and large ships! It's pretty obvious that neither of these are the cause of the outages that Andy is seeing here though. Regards, Greg On Tue, 2003-10-28 at 11:30, Jeff King wrote: > > Greg: > > In that light, wouldn't circular polarizaion be the best solution since any > reflection of a CP polarized changes its twist, hence giving a CP antenna 20+ > db rejection of the 1st order multi-path? > > I guess the downside here though is I'm not aware of any off the shelf CP > antennas with 24 dBi gain. > > Still, even though it is fairly clear what the problem is, we still do not know > the link margin to begin with. It sounds like it is just a 1megabit link on the > peaks, hence I'm suspecting we have a poor link margin to begin with. > The "quick solution" may be as simple as replacing the WAP11's with some > AP's/clients with hotter recievers. Otherwise it will be up and down the towers > and adding a diversity dish possibly. > > Quoting Greg DesBrisay <[EMAIL PROTECTED]>: > > > Countermeasures can include (1) using horizontal polarization instead > > of > > vertical polarization (vertically polarized waves flip and change > > phase > > by 180 degrees when reflected from a conductor, horizontally polarized > > waves don't change phase when reflected), and (2) positioning your > > antennas so the area where signal reflects off the water is over -- general wireless list, a bawug thing <http://www.bawug.org/> [un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless
