Single or Multi channel?

[anotherenduser]

I read/hear so many things about single channel being better than multi channel and vice versa..... What is the truth behind this? Which is the best way to deploy a network. If this has been covered before I'm sorry.....

[SYNACK]

Everyone does multichannel, meru does single channel, meru does work but everyone else in the market has chosen to,go the other way. Both ways work but the majority go with multichannel.

[njc235]

Both work well when deployed in the right way in the right environment. Sometimes multi cell os a better choice, other times single cell is better. Many factors need to be considered some of which are;

Throughput requirement

Concurrent connectivity

802.11ac viewpoint

Building(s) layout and construction

Legacy experience (people often choose one or the other because it's all they've ever know)

 

A pre-sales discussion to understand the differences and the choices is as important as the pre-installation RF survey.

 

HTH

[wifispray]

The reality is like "SYNACK" says only Meru (and extricom) work with a single channel architecture. All other vendors are not focusing on this, through choice and not limitations in architecture.

 

Without testing each vendor it is hard to say in your environment which will work best. Single channel is best suited to smaller and less dense environments and where the RF environment doesn't change very much (unless you are willing to look at channel layering which adds cost). Multi channel or Microcell is more adaptive and has more build in features to handle density environments (like RRM and using multiple channels).

 

There is always much discussion in the industry about this, but to date none of the major vendors (Cisco, Aruba etc) have decided to move down the single channel route. Only they can comment as to why?

[njc235]

There is always much discussion in the industry about this, but to date none of the major vendors (Cisco, Aruba etc) have decided to move down the single channel route. Only they can comment as to why?

 

I believe the Meru model is protected by patent and cannot be copied. The Extricom single channel or 'Blanket' model requires each AP to be wired back to an Extricom controller rather than any switch on the network.

[psydii]

Of course, worth noting that the typical cell architecture doesn't scale in a 11ac world where you have to avoid DFS channels.

 

I suspect that everyone will be offering 'channel blanket'-like solutions over the next few years, in much the same way that beam forming is no longer a unique selling point for ruckus.

[wifispray]

Of course, worth noting that the typical cell architecture doesn't scale in a 11ac world where you have to avoid DFS channels.

 

I suspect that everyone will be offering 'channel blanket'-like solutions over the next few years, in much the same way that beam forming is no longer a unique selling point for ruckus.

 

Everyone has issues with DFS channels single channel or multi channel.... Hopefully OFCOM will free up more spectrum to mitigate this issue.

 

I don't think everyone will be offering channel blanket like solutions in the future, what makes you say that? Beamforming is an interesting one... Ruckus choose to focus there product on this (they have the best AP at L1). Now with 802.11ac this becomes standards based so no longer single benefit to Ruckus, but they maybe understand this better than anyone. Fact is Beamforming is very important for the next phase of 802.11ac wave 2 where we have MU-MIMO.

[psydii]

Because there are so few channels available you can't get anywhere near 1gbit performance in real world deployments without accepting that cohabiting channels will be able to hear each other very loudly. At which point you need all the single channel tricks you can muster to mitigate the impact.

[Net-Ctrl]

I suspect that everyone will be offering 'channel blanket'-like solutions over the next few years, in much the same way that beam forming is no longer a unique selling point for ruckus.

 

Not strictly true, there are different fundamental categories of beamforming, including chip-based beamforming and antenna-based beamforming (e.g. BeamFlex). Most vendors are limited to offering whatever their chip vendor has included in the Wi-Fi chipset. By contrast, Ruckus can support both methods simultaneously, so there’s no need to worry about choosing one or the other.

 

They guys at Ruckus have produced some pretty good blogs and whitepapers about beamforming:

Beamforming Basics for the Wi-Fi Challenged - The Ruckus Room

http://theruckusroom.typepad.com/files/beamforming_response_doc_final_2013-1.pdf

 

Edit: There is also a video that is really good from Steven Glapa at Ruckus on this subject: https://www.youtube.com/watch?v=jn0_QzJHahk

Edited July 23, 2014 by Net-Ctrl

[wifispray]

But on the flip side with cohabiting channels you are open to more of the effects of co-channel interference or for a better work co-channel congestion. The system would need to be more TDM like to fully achieve this, i don't believe any vendor has achieved this yet...

[wifispray]

Not strictly true, there are different fundamental categories of beamforming, including chip-based beamforming and antenna-based beamforming (e.g. BeamFlex). Most vendors are limited to offering whatever their chip vendor has included in the Wi-Fi chipset. By contrast, Ruckus can support both methods simultaneously, so there’s no need to worry about choosing one or the other.

 

They guys at Ruckus have produced some pretty good blogs and whitepapers about beamforming:

Beamforming Basics for the Wi-Fi Challenged - The Ruckus Room

http://theruckusroom.typepad.com/files/beamforming_response_doc_final_2013-1.pdf

 

Edit: There is also a video that is really good from Steven Glapa at Ruckus on this subject: https://www.youtube.com/watch?v=jn0_QzJHahk

 

Very true. However, The thing with Ruckus' solution is client support and moving to more than just downlink. I think standards will win out...... they always do. But Ruckus stand a great chance at leading the pack.

[White_Fi]

To improve shared medium systems such as WiFi you need to get that client on and off the link as fast as possible and move onto the next client. The faster you can do that the less likely of a collision. It's a probability game and higher data rates vastly reduce the chance of collision.

 

802.11ac is a big game changer for this and the chances are SCA, regardless of SCA MAC alteration techniques being used, will working better once a greenfield VHT environment is achieved (years away from that).

 

Adaptive antenna array (Ruckus BeamFlex) is all about providing a better SNR to the client.. The better the SNR the better the clients data rate.. in a nutshell.

 

802.11n and 802.11ac Chipbased beamforming (TxBF) requires that the multiple streams of data be exactly the same... Great.. But now you cant do spatial multiplexing whilst TxBF is going on.. So that 3x3:3 laptop we are all specking is now using only a single stream rather that its SS capabilities.

 

Since Ruckus "Form" shapes of RF energy a positive way for all clients at the antenna level they are able to achieve spatial multiplexing and "Beamforming" at the same time.

 

All vendors have great bells and whistles but Ruckus, to my knowledge, are the only guys in the marketing taking that chipmakers reference design and throwing it in the bin then saying

"Let's make our own antennas and not follow this design guide... We are an antenna company and our WiFi rocks!"

 

The results have show some pretty awesome WiFi wold wide.

 

The video on beamforming above is well worth a watch to clarify on the underlying technologies.

[wifispray]

As with everything, over marketed on the actual performance gains, but at PHY Ruckus are currently marginally better than all...... Every vendor has a benefit and short coming.... Its finding which ticks more boxes than another based on your requirements.