MIT scientists develop groundbreaking new WiFi that's three-times faster

[microchip]

Scientists at MIT claim to have created a new wireless technology that can triple Wi-Fi data speeds while also doubling the range of the signal. Dubbed MegaMIMO 2.0, the system will shortly enter commercialisation and could ease the strain on our increasingly crowded wireless networks.

http://www.msn.com/en-us/news/techn...hats-three-times-faster/ar-BBw0beS?li=BBnbcA1

 

[thiggins]

Here's the MIT article.

The basic concept is "distributed MIMO", with phase coordination among multiple APs.

 

[System Error Message]

So its basically MU-MIMO + mesh wifi.

 

[alphamatter]

Ha! I was just reading that here:
http://www.zdnet.com/article/wi-fi-...17cfd61&bhid=20028888122161025002408423065352

Calling it MegaMIMO 2.0 and having all the access point running this, work with each other without stepping on each others spectrum usage.

 

[sfx2000]

So its basically MU-MIMO + mesh wifi.

No, it's not that...

It's a bit different, and has nothing to do with MU or Mesh - it's coordinated MIMO across different AP's...

Devil's in the details with it, and it'll be hard to commercialize - interesting research, but MIT isn't the only one working on things like this.

 

[sfx2000]

In any event - we're getting so close to the Shannon Limit, things like this are incremental at best...

We keep finding ways to push it, and with WiFi, we're actually closer to that limit than LTE or WAN Fibre...

And once that limit is hit, it's all just noise...

That's the practical limit that the startup I was part of hit with Mesh, and this is the same limit that hits MU...

MU gets to a point where the channel is totally saturated - no more coherent energy can be put into the channel and extracted at the other end, as Eb/No is a hard limit - and that's what worries me about MU - it's not the local BSS, it's the adjacent BSS that can suffer...

Things like MU, QAM1024, this MegaMIMO - these are all technologies on the long tail - and they're out in the parking lot picking up nickels so to speak... spending a lot of time to get little benefit within the construct we consider WiFi with backwards compatibility going all the way back to 802.11 (pre-B) and 802.11a.

My sidebar thoughts here is that we've hit a practical limit with CMSA/CA in WiFi that perhaps we need to go to a scheduled MAC, where we do collision detection rather than avoidance, but this isn't backwards compatible with WiFi, but in the time domain, does help...