I believe a 1 x 2 antenna arrangement is possible, ie 1 TX/Rx antenna at one end of a link and 2 antennas/chains at the other.
If I’m right, at the two-antenna/chain end is it better to have both antennas TX and RX or one TX and one Rx. And in the former case does the card divide the specified Tx power between the cards. Ie with 2 x 23dB antennas is there a potential antenna Tx gain of 46dB. Presumably this assists Rx but both antenna connectors on default Tx simultaneously would put a lot of stress on the card’s circuits and components.
Isn’t it wonderful that a great new protocol like N is developed by a small group of experts, put into effect by a slightly larger group of experts who have the time and know-how to comprehend the technical papers and leaves the rest of us admiring it from a distance and wondering how it’s done?
The cards are designed to be run 2tx and 2rx. It isn’t pushing their limits by allowing them to work as designed. That being said, the antenna configuration must take in account the dual stream setup. Two vertical polarity antennas must be set far enough apart horizontally (kind of difficult) or vertically separated (much easier) to allow dual V or dual H use. This way doesn’t give 150mb @ 20MHz, however it makes single data stream connections more reliable.
Most prefer dual polarity as it works a bit better, a bit faster.
One tx and one rx in an 802.11n system is comparable to throwing away money. There is no benefit to running a 1x1 with two amps, two antennas, etc, splitting the tx and rx. You get the quirkiness of 802.11n working better with a clear band without the error correction of a dual chain config.
“Figure 3 depicts MRC operation graphically for a 1x2 channel. In this example, the two channel gains have magnitudes of 3 and 2. With expected noise power 1, these gains correspond to SNRs of 9 and 4, given that a signal’s power is the square of its magnitude. The MRC receiver scales each antenna’s signal by its magnitude, normalized to the total; delays the signals to a common phase reference; and then adds them. The result has magnitude v13, and the normalized weighted sum of noise still has expected power 1. The combined signal thus has a resulting sum SNR of 13.” (“802.11 with Multiple Antennas for Dummies”, Daniel Halperin, Wenjun Hu, Anmol Sheth and David Wetherall. University of Washington and Intel Labs Seattle)
Sorry, highly technical and almost useless. Are you working on the 802.11n protocol or using the devices?
Reminds me of a amateur radio test booklet I was looking at the other day. All kinds of info in the book, on the test, none of if practical. VSRW questions that don’t bring up the fact that you will hardly ever see a VSRW better than 1.5.
I’ve tried to help you out on some stuff and I don’t really see where anyone else has. Your commentary of how two 23 dBi antennas equal 46dBi shows an underlying lack of understanding of the field. Your attitude towards those few people trying to help might have something to do with a lack of responses by those that frequent this forum.