At 9/25/2011 06:31 PM, Robert Canary wrote:
>Thank you, that's the most (and the best) info I have gotten on the 
>UBNT from anyone.
>
>Could you explain what the "MCS is, and why one would use it?

MCS means "modulation and coding scheme".  You can see the list of 32 
MCS options in this Wiki article:

http://en.wikipedia.org/wiki/IEEE_802.11n-2009

Note that MCS0 through 7 are single-chain ("spatial streams"), 8-15 
are the same modulation but dual chain MIMO, etc.  So in theory MIMO 
lets you have the same amount of capacity for each chain, so MSC 31 
(260 Mbps) is four times MDS 8 (65 Mbps).

What each MCS option specifies are two variables.  The modulation 
tells how many bits/baud, with MCS0 being BPSK (one bit at a time) 
and MCS 5-7 being 64QAM (6 bits at a time).  The coding tells how 
much forward error correction is applied, with payload as a ratio of 
total bits.  So MCS 0, 1, and 3  are 1/2 coding, so the payload is 
half the actual transmitted bit rate.  This amount of parity allows 
for a pretty high received error rate.  MCS 7 uses 5/6 coding, so it 
tolerates a much lower (but non-zero) received raw error rate.  So if 
the link is a bit noisy and you're pushing MCS7, just going to MCS6 
(3/4 FEC) might work, but the throughput falls from 65 to 58.5 Mbps 
(in 20 MHz, using 800 ns. guard times).

Licensed radios are generally hard-set to one modulation scheme, 
often 64QAM or even more aggressive, like 256 QAM, to meet FCC 
efficiency rules.  So they need a strong signal or they crap 
out.  Recent FCC rule changes (or proposals; I don't recall if 
they've been approved yet) will allow fallback to a lower rate for 
some small percentage of the time.  This is viewed as radical in the 
traditional licensed world; it's an everyday technique in the 
unlicensed UBNT or MikroTik world.  So these cheap radios are in some 
ways more powerful than expensive licensed ones!

>Robert Canary
>OCDirect Electrical-Datacomm
>(866) 594-0786 Fax
>(270) 955-0362 Voice
>
>----- Original Message -----
> > At 9/25/2011 02:23 PM, Robert Canary wrote:
> > >Keeping a link active versus maintaining throughput under divers
> > >conditions is two different things. For the money paid I would go
> > >with something like Alvarion.  But then again, after 12 years, I
> > >would not invest big dollars in CPE or Access points.  Only in the
> > >backhauls and infrastructures.
> > >
> > >The only reason I have not went UBNT, I have found much feed back on
> > >how they deal with interference, I like Frequency Hopper (FH) they
> > >keep a decent link through the most divers environments.  But how
> > >does UBNT deal with interference?
> >
> > UBNT uses chips that are essentially software-defined radios.  They
> > implement the 802.11 G, A and N modulation.  G and A are a fairly
> > simple OFDM.  -N is an OFDM with MIMO capability and some additional
> > features.  FH and DS are both older spread spectrum techniques; OFDM
> > is wideband, but not really spread spectrum.
> >
> > The N specs in particular (which work on both 2.4 and 5 GHz bands)
> > include a lot of modulation options ("MCS").  So you can select the
> > modulation that works best on the link in question, and choose 5, 10,
> > 20 or 40 MHz channels.
> >

  --
  Fred Goldstein    k1io   fgoldstein "at" ionary.com
  ionary Consulting              http://www.ionary.com/
  +1 617 795 2701 



--------------------------------------------------------------------------------
WISPA Wants You! Join today!
http://signup.wispa.org/
--------------------------------------------------------------------------------
 
WISPA Wireless List: wireless@wispa.org

Subscribe/Unsubscribe:
http://lists.wispa.org/mailman/listinfo/wireless

Archives: http://lists.wispa.org/pipermail/wireless/

Reply via email to