Advertised claims for "multi-polarized" antenna

This antenna claims to be better for picking up signals around obstructions. Any truth to this? Cf Multi-polarized antennas.

formatting link
sounds a little bogus to me?

Reply to
Loading thread data ...

Greater tree penetration?

It's bullshit.

Reply to
me here

These are all the same company:

Bogus. The general lack of specifications, models, and test results, should offer a clue.

Antennas do not create RF, they only direct it. If you split your RF power into multiple polarizations at the access point, but your client radio only has a simple antenna that can use only one polarization (vertical, horizontal, right-hand circular, or left-hand circular), you have lost most of your power into useless polarizations.

For indoor use, it's even worse. 802.11 has problems dealing with reflections and multipath resulting in frequency selective fading. By splitting the signal into multiple polarizations, one would think that the client end would work better because at least one of these multiple polarizations might be usable. However, that's not what happens. Instead, multiple polarizations simply create more opportunities for more reflections. It's not like 802.11n (MIMO) spacial diversity, which requires seperate antennas and multiple orthogonal (non-interacting) signals.

Reply to
Jeff Liebermann

Meanwhile, at the alt.internet.wireless Job Justification Hearings, Jeff Liebermann chose the tried and tested strategy of:

Does this apply even if you have multiple clients with differing polarisations? I assume you're making a distinction between splitting a single RF feed into multiple polarisations, and connecting a differently polarised antenna to a diversity connector.

Yes, that's what I would have assumed. So, the extra paths the signal takes outweighs having the correct polarisation?

Reply to

Sorta, kinda, maybe, etc. As with anything concerning RF, it's not very simple.

If I take a mythical lossless power divider, divide the tx power by four, and send each one to a different antenna each with a different polarization, the typical client radio will react differently to each of the 4 polarizations. I'm going to skim over quite a few basic concepts. Bug me if you need detail.

I'll assume that you're using a typical laptop, with tow PIFA antennas mounted horizontally near the top of the screen. Note that this arrangement is horizontally polarized. If the AP antenna was vertically polarized, theoretically such cross polarization wouldn't work. As most users have observed, it does work. Two reasons. The loss due to cross polarization between vertical and horizontal is only significant near exactly 90 degrees cross polarization. At most other angles, it's far less. The other is that much indoor propagation is via reflections, not the direct path, where the polarization is uncontrolled and almost random.

With circular polarization, the isolation between right hand (RHCP) and left hand (LHCP) is quite large. If the access point were using one sense, and the client using the other, it would work only with odd numbered reflections, which reverses the sense. Assuming both ends were using the same sense, there would be no polarization mismatch losses, but not for odd numbered reflections, which reverses the sense.

To the best of my knowledge, there are few client radios using CP helix antennas, so neither of the above scenarios are realistic. However, were we to use the previously mentioned horizontal antennas on the laptop, with some CP antennas on the AP, then there would actually be a tiny benefit. There would be a 3dB polarization mismatch loss between the CP and horizontal polarizations. That's at almost any orientation of the laptop antennas, eliminating the deep (theoretical) cross polarization null when everything is linearly polarized. (This is why GPS is CP, so that it works at any orientation).

There is also elliptical polarization, which is a combination of linear and circular polarizations. The cross polarization effects are still there, but not as pronounced (and not as easily calculated).

So, what does all this have to do with the mythical multi-polarization antenna? With a common PIFA laptop antenna on the client, if you split the power 4 ways into 4 polarizations, at least one of the 4 polarizations will exhibit substantial loss to the client antenna. This makes the splitting of polarizations LESS effective than if the AP and client used the same polarization. In addition, it is difficult (or impossible) to maintain the same ratio of polarization diversity over the entire hemispherical or torus antenna pattern (missing from the data sheets), where the signal might be vertically polarized in one orientation, and horizontal in another. More simply, the polarization is probably not very consistent.

Lets's try an example. If I split the power into 4 polarizations (vertical, horizontal, RHCP, LHCP) with various client antennas, the

*ADDITIONAL* losses (in dB) introduced by polarization mismatch would be something like:

Vert Horiz RHCP LCHP Vert 0 -30 -3 -3 Horiz -30 0 -3 -3 RHCP -3 -3 0 -30 LHCP -3 -3 -30 0

If you assume the columns are the AP and the rows are the client, there is no single client antenna polarization, that does not exhibit additional loss from the multipolarized contrivance. Note

0dB doesn't mean no loss as the signal is already -6dB down from the original power level due to splitting the power 4 different ways.

Extra paths are only beneficial in MIMO, but only if they are orthogonal (non-interfering) between each other. MIMO takes considerable effort to craft the modulation so that each signal minimally interferes with the others. This requires separate transmitters, receivers, and antennas. You don't get the speed benefits of MIMO with a single antenna and single xmitter.

For non-MIMO (802.11g), extra paths cause some inter-symbol self-interference and frequency selective fading. They are only are useful if diversity reception is aggressively employed.

Reply to
Jeff Liebermann 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.