What's new

Practical use for 4x4 mimo WAP in 2.4ghz

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

Cloud200

Senior Member
Just wondering out loud.
Did any manufacturer ever make a 4x4 2.4ghz radio?
I know complexity goes up exponentially going from 2x2 to 3x3 and I assume the same applies to 3x3 to 4x4.
Would there be any practical use for it?

I know in the enterprise space there are 4x4:3 radios where they choose the best 3 of 4 to communicate with.
I am talking about true 4x4:4
 
Do any 4x4:4 (802.11n) clients exist?
 
Do any 4x4:4 (802.11n) clients exist?
Build it and they will come?
I was just thinking from an engineering standpoint what the benefit would be going from 3x3 to 4x4. Would there be too much noise from the other antennae to make a difference in anything but the most optimal of conditions?
 
From an engineering standpoint, I assume if bandwidth is a priority, then it is better to choose 5Ghz.

I see 2.4Ghz 600Mbit happening only when it is super-cheap to manufacture. It seems like a wasteful investment otherwise.


This is just my guess though. :)
 
Just wondering out loud.
Did any manufacturer ever make a 4x4 2.4ghz radio?
I know complexity goes up exponentially going from 2x2 to 3x3 and I assume the same applies to 3x3 to 4x4.
Would there be any practical use for it?

I know in the enterprise space there are 4x4:3 radios where they choose the best 3 of 4 to communicate with.
I am talking about true 4x4:4
4x4 in 2.4 GHz is in the next generation QCA based AC2600 class routers. 800 Mbps link rate with 40 MHz bandwidth.

These are also supposed to support MU-MIMO at shipment.

Good luck finding MU-MIMO STAs.

Closest thing you will get to a 4x4 client devices will be bridges.
 
Would you be able to test MU-MIMO once they arrive?
I assume you would need a combination of 1, 2, and 3 radio clients all testing at the same time.
Then a few comparative runs with devices running AC1200, AC1750, etc . . .
 
Would you be able to test MU-MIMO once they arrive?
I assume you would need a combination of 1, 2, and 3 radio clients all testing at the same time.
Then a few comparative runs with devices running AC1200, AC1750, etc . . .
MU-MIMO testing depends on finding clients. I don't have any. I will be looking to vendors for help.
 
4x4 in 2.4 GHz is in the next generation QCA based AC2600 class routers. 800 Mbps link rate with 40 MHz bandwidth.

These are also supposed to support MU-MIMO at shipment.

Good luck finding MU-MIMO STAs.

Closest thing you will get to a 4x4 client devices will be bridges.
Tim will these 4x4 in 2.4 GHz devices have to follow the 20Mhz co-existence rule?
 
They need to be 802.11 compliant. So yes.

That said, from what I see, 20 MHz coexistence is more a suggestion than a rule. Manufacturers routinely allow it to be disabled.

I stopped checking for it awhile ago because I got fed up with how often it was ignored or allowed to be disabled.
 
I've recently come across it with newer wireless devices. I wonder if the FCC certification is now enforcing compliance in new firmwares.
 
For MU-MIMO, it absolutely makes sense, since you lose a stream during multiuser operation.

It might make some limited sense in some cases without MU:MIMO, as you have higher potential signal gain in MIMO operation the more spatial streams you have. Also potentially better explicit beam forming performance, though it sounds like most radio makers are only doing EBF on 5GHz (though maybe Broadcom is doing it on 2.4 and 5GHz? I am pretty sure QCA is 5GHz only).

For the added expense and complexity, without MU:MIMO it likely does not make any sense to do. However, you can say the same about 5GHz band where it doesn't really make sense to make more than 3 spatial streams other than MU:MIMO operation. With MIMO signal gain and EBF, I'd imagine you see some fairly reasonable gains moving from 2 to 3 spatial streams and significantly less gain moving from 3 to 4.
 
MU-MIMO is part of 802.11ac. So 5 GHz only unless someone has gotten really creative...
 
MU-MIMO in 2.4GHz - practically speaking - no... it needs explicit client support to work.

Will the Marketeers ask the Engineer's to do it? - Probably yes, as most WiFi chipsets support both modes, and it'll be a checkmark on the back of box specs...

Just saying - we're already seeing 802.11ac attributes being pushed down into the 2.4GHz space to support non-standard TurboQAM...
 
For MU-MIMO, it absolutely makes sense, since you lose a stream during multiuser operation.

Personally - having been part of the LTE development process - MU-MIMO makes a lot of sense with SmartAntenna's in a WAN perspective - for 100's of users - consider that a typical 1.9GHz LTE/3G/2G cell can support in a decent manner over 600 users (200 per sector), then MU-MIMO can make sense...

It doesn't in a "small cell" environment with limited number of users...

In the WLAN environment, MU-MIMO isn't going to provide near the throughput gains that laymen expect, basically because the of the constraints needed for it to actually work...

So for 802.11, I'll go on the record - MU-MIMO will be a disappointment to most at a very high cost...

That being said - there's still a lot of headroom in 802.11ac - truly wide channels, e.g. 160MHz, 4-stream's with SU-MIMO, better implementations of SU Beamforming perhaps with adaptive antenna arrays...

Just my 2 cents...

sfx
 
Just wondering out loud.
Did any manufacturer ever make a 4x4 2.4ghz radio?
I know complexity goes up exponentially going from 2x2 to 3x3 and I assume the same applies to 3x3 to 4x4.
Would there be any practical use for it?

I know in the enterprise space there are 4x4:3 radios where they choose the best 3 of 4 to communicate with.
I am talking about true 4x4:4

FWIW - the Linksys WRT-1900 is a 3-stream, 4 Tx/Rx radio solution, which probably explains why it performs the way it does...

With the client mix, it's diminishing returns, but all clients will benefit somewhat for more radios and streams... e.g if one has an AP with 3:3*3 and a single radio client (1:1*1), there's some precoding that is done on the transmit side that can improve signal robustness, and the three radio receive chains will increase gain there...

So a bit better range, and at a given range, perhaps a bit better performance..
 
Problem is on the client side (handheld, notebook).
Antenna sizes and spacing.

Notebooks, not so much - lots of room and space for good antennas... and generally they tend to be used in a single orientation..

Handhelds/SmartPhones - there we're in perfect agreement... small form factors, and landscape vs. portrait mode - and most designers need to also consider bluetooth, GPS, and multiple WAN bands, and the WAN bands (GSM/CDMA/LTE) tend to take priority for antenna placement/orientation...
 
Notebooks ... yes. But many/most can't or don't put antennas in the LCD half of the laptop. Too hard, hinged, etc.
 
Notebooks ... yes. But many/most can't or don't put antennas in the LCD half of the laptop. Too hard, hinged, etc.

I haven't looked at a ton of tear downs recently, but almost all laptops I've ever pulled apart have the antennas in the LCD area, generally wrapped around the bezel. I think I've pulled apart zero laptops where the antennas where in the keyboard area. Maybe trends have changed in the last year or two.
 
MU-MIMO in 2.4GHz - practically speaking - no... it needs explicit client support to work.

Will the Marketeers ask the Engineer's to do it? - Probably yes, as most WiFi chipsets support both modes, and it'll be a checkmark on the back of box specs...

Just saying - we're already seeing 802.11ac attributes being pushed down into the 2.4GHz space to support non-standard TurboQAM...

Yeah. From what I have seen, most manufacturers are extending SOME of 11ac's specifications in to the 2.4GHz space. Not just 256QAM, but also a number of other 11ac "standards", like low parity error checking seems to be something that is more standard in 2.4GHz client and router chipsets that have 11ac for 5GHz. It was purely optional in older 11n chipsets and rarely adopted. It seems much more common now.

Some manufacturers do appear to be doing EBF in 2.4GHz, also IBF. It was optional in 11n and almost no one did it then. Now with 11ac it seems like a large number of basestation 11ac 5GHz chipsets are doing EBF (and some IBF too it seems like) and some of the 2.4GHz chipsets are also doing at least IBF and some EBF as well (the little I've been able to tease out, Broadcom has at least some 2.4GHz radios that are doing IBF and possibly EBF as well, though that isn't clear. Marvell has several 2.4GHz that are doing IBF and EBF).

To the best of my knowledge, the upcoming Broadcom radios that are 2.4GHz 4:4 DO support MU:MIMO. Also the upcoming 11ac MU:MIMO chipsets for clients from QCA and Broadcom I think support MU:MIMO in 2.4GHz, but I could deffinitely be wrong about that.
 

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top