Forum

  • When we have “conventional radio systems”, we would like to have all reflected signals be exactly in phase with the “main signal” at all times. Sadly this only occurs a certain percentage of the time. For the rest of the time, the reflected signals will be “adding a little bit” or “removing a little bit” and “distorting a little bit”, the main or incident signal. Multipath is not usually a good thing in these circumstances.

     

    When we use .11n, we want to try to make sure that there is “no exact overlapping of incident and reflected signals”.

     

    Why ? Well back to that little guy who is  stuck in the box. If someone speaks German and someone speaks Italian and someone speaks French at EXACTLY the same time, he would have a very hard time discriminating between the signals. But if there was a very slight delay when each signal reached it’s peak [ I have to be very careful here, as this is a very, very rough analogy, and it doesn’t work exactly like this ] he would stand more of a chance of understanding what was being said.

     

    The key to all of this is that we should not be able to predict an exact phase relationship between one wave and another. We can do this by means  of multipath or, as Gene says, by means of polarization.

     

    How can a vertically polarized antenna receive a signal from a horizontally polarized antenna ? That sounds wrong.

     

    The key is in what is known as the Antenna Cross-polarization Discrimination Ratio or XPD ratio.

     

    In an absolutely perfect situation, a vertically polarized antenna would not be able to receive a signal from a horizontally polarized antenna, but in reality, they can [ and it works vice-versa ]. [ I won’t go into why this is at this stage ]. With an antenna with a low XPD ratio, we would see more of the opposite [ or x-pol signal ] than from an antenna with a high xpd ratio.

     

    Ruckus must be using some sophisticated processing in their systems, as usually [ non-802.11 ]  the introduction of a cross-pol component merely shows up as a reduction in Co + No/No

     

    Gene do you have [ provided it’s not top secret !! ] any more info about the polarization business ? You also had a nice presentation about some of this stuff from the road show. Is that still available anywhere ?

     

    Dave

     

     

Page 1 of 1
  • 1