• Hello,
    I've done developing this feature to AP and now turnning to STA development.
    I've encountered some difficulties with the STA's scan exemption

    When needs to calculate the ActivityFraction, T-active is defined as the total duration of trasmitted MSDU's and received individually addressed MSDU's guring the previous T-measur-active seconds (which is DelayFactor * OBSS Scan Interval)
    This demand is pretty heavy, does anyone knows a manufactur who implemented this (or exemption request )?since i've tried Intel's STA and there was no such request from them (checked with a sniffer).

    According to the standard, STA is exempted from scans only if received exemption from it's associated Ap and whenever its ActivityFraction is lower then ObssScanActivityThreshold/1000.
    i once saw a Broadcom paper which says their are using the value of 300 for their ObssScanActivityThreshold variable (default and maximum value), which means that when their ActivityFraction is above 3%, they make the scans.
    What is the logic behind that (why such a low value)?
    Why should the STA make the scans when it needs to work hard and not leave it to other STA's who work less, is it because "if your using the AP so much, at least do something for it"?


  • I this a ddwrt specific question?

    Do you have an 802.11 sub-clause to reference ?

    Or is this all gobbledegook?

  • Sure,
    In Draft P802.11-REVmb/D09.0, May 2011 it's in sub-clause 10.15.6
    Anyway, the sub-clause name is [b]Exemption from OBSS scanning[/b]

  • This is not my area of expertise, but I'll take a stab at it.

    The 3% value may be due to the fact that using anything over it defeats the purpose of trying to share the spectrum.

    Do you have any experimental data of your own that might explain their choice? I would bet that by 5 or 6 % performance for everything else has gone to hell.

    I don't know if you can request an interpretation on a P document. You may have to join TGn to get an answer at this point in time.

Page 1 of 1
  • 1