[SGVLUG] Algebraic wifi

[Christopher Smith]

On Tue, Oct 23, 2012 at 6:04 PM, Braddock Gaskill <braddock at braddock.com> wrote:
> On Tue, 23 Oct 2012 17:01:57 -0700, Doug Vargas <dougvargas at sbcglobal.net>
>
> wrote:
>
>> Saw this via slashdot, the speed ups are simply mind blowing, it'll be
>
>> fascinating to learn more about the math used to compute the packets -
>
> sort
>
>> of like an equational hash sum?
>
> Look into FEC error correction coding and specifically error erasure
> coding.  This stuff has been around since WWII, and I've frankly been
> mystified why it hasn't been applied to Wifi on the packet level since my
> first sketchy wireless coffee shop link 15 years ago.
> It does incur latency and throughput overhead on networks without loss.

Yeah, I think there is more to it than that, although I haven't seen
the real details yet. Wireless communications have in general been
using a combination of FEC and ECC to push things right to the Shannon
Limit for quite some time... so I'm not sure what the big deal is here
other than perhaps additional efficiencies that can be gained by
employing the approach at layer-3 (and even then, there are plenty of
examples of that kind of approach out there).

-- 
Chris