[Opinion] 2×2:2 MIMO in Smartphones

So, Broadcom announced a 2×2:2 (867Mbps) chip for mobile devices, which of course generated lots of excitement.

Let’s see: 867Mbps Wi-Fi, you say?

  • Can you sustain such speed with your home Internet connection?
  • Have you seen hotspots offering such a high-speed Internet?
  • Do you have a RAID-enabled NAS at home for file transfers, since even the fastest SSDs do not allow for such read rates? [UPD] Well, clearly got carried away and confused MB and Mb here [/UPD]
  • And if the fastest SSD cannot read that fast, what makes one think their tiny phone will be able to write faster onto their embedded flash (or SD card)? The data I have is 35-65Mbps write rates on average.
  • Two antennas means 2x+ power usage of the WLAN module. Can you imagine effect on the battery life?

So, what is the point, then, you say?

For small burst transitions (downloading web pages) that can go directly to memory – this higher speed could be quite useful, as transmission takes less air time. Less air time means better client density on one AP. Less air time also means less power used, and radios don’t stay on for too long (Broadcom claims 25% overall reduction in power use). I am very interested to test this live and see if there is a noticeable difference over “simple” 1-stream 802.11n/ac in real-life scenarios – there’s much more to 802.11 protocol than sending data frames.

What could really work, is MRC and STBC with 2×2:1 or even 1×2:1 mode

  • MRC provides better receive sensitivity for the decoder, using math to combine signals from both antennas and extract ‘cleaner’ source data. This can generate ~3dB extra receive gain.
  • STBC provide better chances for MRC receiver on the Tx side, by transmitting multiple copies of the same data stream from both antennas in a way (math, again) that MRC receiver can utilize them even better, with potential for another 3dB gain.
  • As a result, one gets better rate-over-range (esp, when both AP and client support both technologies), which is very essential for a compact-size device that has it’s WLAN antenna largely covered by a human hand.

So, just like with 802.11ac itself, 2×2 in a phone is not about speed, but rather coexistence and connection quality. And I still believe it’s overkill – I would stop at 1×2:1, probably, for battery life reasons. Expect another wave of spec-selling, just like 800-1000Mhz LCD SmartTVs 🙂


4 thoughts on “[Opinion] 2×2:2 MIMO in Smartphones

Add yours

  1. think the bigger point is getting these speeds pervasively in a business which is the next step. The WLAN itself is an issue because how can you implement when you are down to 2 useable channels due to DFS. so home use one thing but business use case – not with traditional approach to wireless. its highly misleading and is just physics


    1. Eric, thanks for commenting.
      Of course everyone would want higher speeds and less airtime utilization to enable better coexistence and higher capacities. That’s what 802.11ac is about. The only question is, are you getting what you’re paying for? What to expect from a 2-stream mobile phone (be it home or enterprise WLAN)? Will the difference (compared to 1-stream .11ac phone) be noticeable? Will it be worth it in terms of battery life (another important Enterprise consideration)?
      BTW, yes, the use of 80Mhz+ channels is fairly limited, but 802.11ac has much better coexistence with 20M/40Mhz channels, so you can essentially plan for those, if you do your planning correctly. Or use DFS. 🙂
      What part, exactly, have you found misleading?


  2. I don’t see a lot of 80 MHz channel possibilities happening in general, and certainly not in the enterprise. But, the battery life is actually supposed to be better with these new chips- faster data rate = not as much transmit time, etc- but the radios still need to be powered. We’ll see. I doubt most users will “see” the difference between 11ac and 11n from a smartphone, but I like the FM tuner add.


    1. Lee, Broadcom claims 25% less energy despite two radio chains, exactly due to less airtime.
      But you never know under what test scenario they have got it. So it can be anywhere between -25% and +100% depending on the usage pattern. And then, it still depends on how bugged the drivers will be 😉


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog at WordPress.com.

Up ↑

%d bloggers like this: