Rapidly rising performance requirements on enterprise and carrier Wi-Fi networks dictate squeezing every available Mbps out of infrastructure gear — naturally driving increased interest in using any and all advances in RF technology.
One in particular, so-called transmit beamforming (commonly abbreviated as TxBF), is getting much more attention these days. While this is a potentially useful tool, be careful not to be fooled by vendor claims. As always, the devil is in the details.
Since discussing RF technologies with precision can get complicated, a careful, step-by-step search for the devil behind TxBF really requires the bandwidth of a proper whitepaper rather than a simple blog entry. For those interested in the technical hows and whys, we’ve published the step-by-step story in just such a beefy whitepaper. For the less patient, we’ll summarize the key findings here.
RF BUILDING BLOCKS
Many of the tech nologies available to help improve radio performance come from the broad category known as “smart antennas.” There are many variations on the theme, but the idea common to them all is using more than one antenna on one or both ends of the link to send and/or receive radio signals in a more controlled manner, to increase signal quality and throughput.
There’s now a whole family of multi-antenna techniques that can be employed to achieve RF performance gains in Wi-Fi. In a properly designed Wi-Fi system, all of these tools can be used in combination to maximize results. Click on the chart for more detail
Note that both AA and TxBF are often referred to loosely as “beamforming,” since many in the industry consider the term to mean generically “shaping radio energy in space to focus on the target recipient.” There are fundamental differences in how these two technologies operate; however, yielding very significant differences in the performance improvements they can deliver in the real world.
ASSESSING TRANSMIT BEAMFORMING
Transmit beamforming allows an access point to concentrate energy in the direction of a particular client using signal processing techniques (phasing or timing the signals differently) at the baseband chipset. Explicit client feedback is required for APs to determine the correct phasing for each client.
While a promising potential addition to the RF toolkit, in reality, TxBF is subject to a number of constraints and disadvantages:
No Client Support. There’s simply no way around it. Today this is a complete show-stopper. To achieve any real performance gains with TxBF in Wi-Fi, clients must support the optional feature in the 802.11n standard that provides explicit feedback to the AP about how to do beamforming effectively for each client.
As we show in the figure (right), this feature has zero support in the market today and none on the way in the foreseeable future.
Incompatibility with Spatial Multiplexing. The explanation for this one is definitely best left to our beefy whitepaper, since it requires looking under the hood of how spatial multiplexing in 802.11n really works. The bottom line is that with any commercially practical number of radio chains, it’s impossible to achieve the higher data rates in 802.11n and use TxBF at the same time.
Lots of Self-Interference. With only 3 or even 4 radio chains to work with, TxBF makes very symmetric beam patterns, generally sending as much energy away from the client of interest as it does toward it. This increases self-interference in the multi-AP networks that are critical to success in today’s high-demand-density venues, reducing spectrum re-use and overall system capacity.
Incompatibility with Polarization Diversity. There’s a technical subtlety at work here, too (have we mentioned that we have a beefy whitepaper to cover these things?). The net is that TxBF will fail frequently when used with today’s mobile clients with arbitrary orientation.
Modest Gains at Best. Even when it works, the Wi-Fi chipset engineering community predicts that performance gains in practice will be modest, on the order of 2–3 dB.
ASSESSING ADAPTIVE ANTENNAS
Adaptive antennas — the basis for Ruckus BeamFlex technology — involve manipulating the inherent directionality and polarization of the physical antenna structure itself. This is achieved by electronically switching a subset of a large number of small antenna elements into use with each radio chain for each packet sent. Element selection is optimized client by client, based on achieved throughput, relying on the ACK packet that all clients send as a 100% standard part of the Wi-Fi protocol.
As a result of this unique “layer zero” role in the system, adaptive antennas have none of the operational limitations of TxBF. Specifically, adaptive antennas:
- require no special client behavior beyond mandatory elements in the 802.11 standards (for b, g, or n);
- can be used simultaneously with spatial multiplexing and polarization diversity;
- mitigate interference through highly asymmetric beam patterns; and
- deliver 2 to 3x the performance improvement of TxBF, through better leverage of multipath and statistical optimization techniques.
WHAT'S IT ALL MEAN?
There are circumstances when TxBF will be useful when client support emerges and in combination to adaptive antenna switching. Since it’s available in the next generation of Wi-Fi chipsets TxBF in combination with BeamFlex adaptive antennas, in a form of “BeamFlex 2.0”, offers a best-of-both-worlds solution that yields higher SINR gain with less interference than TxBF alone.
So don’t return to the old omni-antenna reference-design implementations that continue to pollute both the enterprise and carrier network landscape with such mediocre Wi-Fi performance.
Ultimately combining TxBF with adaptive antenna technology will simply deliver the best of both worlds yielding what no other Wi-Fi supplier can provide and every customer wants: PERVASIVE PERFORMANCE.