The Ruckus Room

Demystifying 802.11n

MIMO, WMM,WEP, WPA, 802.1x 802.11a/b/g/e/n/d/i/h – consumers just don’t care and are overwhelmed with all the stuff they think they need to know (cuz vendors tell them it's important). But when it comes to Wi-Fi, consumers really care about three simple things: performance (speed and coverage), price and predictability. 

Well the good news is the IEEE unanimously approved the first draft of the 802.11n standard. 802.11n boosts Wi-Fi’s capacity up to a 600 Mbps theoretical speed and is essential for high-bandwidth applications such as supporting multiple HD streams throughout a home.

But the bad news is that 802.11n, in and of itself, won’t solve how to reliably stream video over the air. So claims that it will be the panacea to support multimedia content mislead already confused consumers. 

Wi-Fi (and the “spectrum” that is uses) is a shared, unlicensed medium. This means anyone can use it anytime for anything.  Sure you get a bigger Wi-Fi pipe with 802.11n but if there’s the slightest amount of noise, your video will suck.  In fact, 802.11n actually might make things worse (but don’t get us wrong, we love N for the big pipeness). But here's why things get worse:

802.11n basically sends, at the same time, multiple Wi-Fi signals out over different paths through the environment (multipath), then puts them together at the other end to deliver a higher data rate.  This process (that I’ve described very poorly) is called (sorry about this) “spatial multiplexing.”  It’s sort of like using multiple APs at the same time and combining them at the receiving end to get a faster connection. 

But this is precisely the problem. Because you can’t control the paths that these Wi-Fi signals actually take, when interference is introduced (and there’s always interference), the Wi-Fi signal can actually become more unstable because it’s bouncing around, making it even harder to recover the different signals. In turn, streaming video over 802.11n could be subject to wild bandwidth fluctuations caused by RF noise and other uncontrollable stuff in the spectrum (I think you know where we’re headed).  Making Wi-Fi predictable is a MUCH harder problem to solve. Ironically wired seems to get this :)

You need to be able to control the paths that signals take and understand the quality of these paths – because we’re talking video here.

If Angelina, when she’s sent from the AP, doesn’t look like Angelina when she reaches the TV, you’ll be pissed.

Our technology makes you not pissed.

WiMAX: Who Needs Metro Wi-Fi Anymore?

WiMAX is just a baby of a technology but it has big eyes on the future and Intel is pushing it like nobody’s business And Intel is one company with the power to make things happen at the desktop. Can you say Centrino? 

They’ve made Wi-Fi a household word and expectation.  Now it seems they want to do the same with WiMAX. Meanwhile legislators are pushing to “open” (read unlicense) some of the WiMAX spectrum such as the 700 MHz band. Where does this leave metro Wi-Fi?

Metro-scale broadband Wi-Fi networks have a huge head start and are being installed like gangbusters from Oklahoma City to San Francisco, Anaheim to Philadelphia with no end in site. Earthlink just jumped into the market in a big way, too, and Google is expected to follow, and is currently deploying a free-access metro-scale Wi-Fi mesh network in Mountain View, California.

The big benefactors of this wave are companies like Tropos Networks and Strix Systems. They build specialized outdoor Wi-Fi access points (often attached to poles) that literally blanket an entire city with a Wi-Fi signal and use very fancy routing software (at least Tropos does) to move traffic from one node to the next over the air.  Tropos has arguably defined, and now clearly dominates, what could be a massive market (and we’re not just saying that because they have our favorites - former FCC Chairman Reed Hunt on their board and Congressman Jack Kemp as an advisor).

But many people think that WiMAX, due to its longer range and higher capacity, will effectively kill metro-scale Wi-Fi and those who sell metro Wi-Fi gear.  Naaaaaaaah.  It’s actually just the opposite.  In fact, it probably means an even bigger positive impact on the valuations of equipment providers in this nascent yet quickly burgeoning market ("burgeoning" was a suggestion by one of our MIT-like RF engineers and not our PR people...can you believe it?).

Think about it.  First of all there are almost 200 million Wi-Fi client devices out there today, and about zero WiMAX client devices.  And Wi-Fi has become (or is becoming) the defacto edge connection technology of choice.  But more important, whether Wi-Fi or WiMAX, to support relatively low powered client devices, such as laptops and handhelds, the “wireless” base stations must be positioned close to the user (or you’d better have a hell of a good device in the home that can listen, and more importantly shout, better and farther).

This points to putting wireless nodes (most likely Wi-Fi at the edge but it could be WiMAX) everywhere. The industry spew for this is “dense cell architecture.” So it really doesn’t really make economic or logistical sense to run a wired IP backhaul connection to each node or to all the cells (backhaul in this context means a single wired connection used to aggregate traffic from other wireless connections in order to provide a route to the Internet or other IP network). Using the RF spectrum to route traffic from node to node to create a redundant mesh is the best way to go. This is really what companies like Tropos are all about - they’ve perfected mesh routing and have chosen 802.11 because it’s by far the most pervasive, simple and low-cost standard.  The best use of WiMAX looks to be providing wireless backhaul from the mesh to the IP network.  Ergo their technology only becomes MORE valuable, not less.

To do this right, carriers will need to make sure subscribers can get on these broadband wireless networks more predictably.  New technology now exists to do just this.  Customer premise equipment that can actually listen farther in a specific direction and listen around interference will help carriers and consumers get on the Internet from their homes without wires.

Now imagine a day when you won’t have ANY lines coming into your home but you’re still completely connected.  The broadband line is now a wireless broadband connection to metro-scale Wi-Fi network.  Over this connection will come digital TV, voice and conventional Internet data services.  And to move all this stuff throughout your house, you’ll use smart Wi-Fi to get signals to your TVs, laptops and refrigerators.  Look Ma, no wires!

What IPTV Isn't and the Profound Scotch Question

Why don't people get it?  IPTV isn't watching TV over the Internet. Om Malik explained this fairly thoroughly in his BLOG.  But it still hasn't seemed to sink in. 

We've spent countless hours trying to explain to people (who SHOULD know) the difference between something like Google video and IPTV.  It's simple. Think of Internet video as merely downloading a file and watching it on your computer in a very small window.

IPTV, on the other hand, is watching (as it is streaming to you in real time) a compressed movie, TV or video content from the carrier's IP network (not the Internet) on a real television. The confusion is that both IPTV and Internet video come into the home over that same broadband (eg. DSL) connection and people don't distinguish between the two. As a result, people don't understand that IPTV is a service they watch, not a file they save. Ultimately the carrier controls the distribution of IPTV content over its own IP network.

That being said, TODAY, people just aren't going to watch (or pay for) TV over the Internet on a big 42" HDTV plasma screen and not all the pieces are in place yet.  But Google, et al are extremely smart, aggressive and smell money. They, undoubtedly, WILL challenge convention by ultimately streaming IP-based video content over the Web so you CAN watch it on your 42" plasma TV.

But for the most part,the Internet was built for data transport and is shared by pretty much everyone on the planet.  This means that predictable performance is almost impossible to guarantee (right now). Sure you can stream a small video file (or download a big video file) and watch it on your computer, but the carriers will never use it as a primary means to deliver real-time video content like movies or VoD. Why?  Because money is too important (if the carriers can't control the quality, they'll be hard pressed to charge for services delivered over it). Ironically consumer-grade Wi-Fi in the home today is in the same boat -an unlicensed band that can't be controlled (we're fixing that though).

But make no mistake, IPTV will do for video what the Internet did for data - more choices, easier access, interactivity, more services, all that.  As soon as people "taste test" this stuff in their homes, there will be no turning back. Once they experience it they will say "I want my IPTV!"  It will fundamentally change our television experience making an explosion of content available, accessible and "interactable."  But we're not there yet and the U.S. market for IPTV is still barely burgeoning.

Meanwhile, ask yourself this: "If my cable company or service provider actually offered IPTV today, what would it take for me to switch?"  This is the multi-billion question being asked by carriers right now.  And it the answer seems to come down to content, flexibility, content, price and content (more on this in a future post it).

On another note...does anyone REALLY drink Scotch because it tastes good?  Be honest with yourself for just one minute. I know you WANT to say yes....but come on. Just a random thought but these are some of the things that confound us.

The CES Hangover

Did you go to CES?  It was wild. Even though we think tradeshows are from satan, this one was worth all the trouble.

One of the coolest things at the show (that provided a glimpse into how things will profoundly change) was the introduction of the NETGEAR/Skype phone. Even though this is a Wi-Fi-only phone (for now), dual-network handsets (that simultaneously support cellular and Wi-Fi technologies) are coming fast and furious from companies like Motorola and Nokia. Here's why they're cool.

So you’re driving home while on your cell phone. You pull into the garage and the cell phone detects your Wi-Fi signal. It makes a connection over your home Wi-Fi network (read “free phone call”) while your still on the cellular network. Once connected, it splices the call from the cellular network to your home Wi-Fi network that ultimately connects to the Internet through your broadband IP connection.  And it does all this without anyone knowing.  That’s going to happen a lot sooner than you think. The point(s) is that these phones accomplish three things:

1. give consumers a more flexible and cost effective way to use the phone (there will essentially be no need for cordless phones anymore)
2. let broadband carriers to steal cellular minutes and cellular money from the cellular phone companies
3. make reliable Wi-Fi in the home a “strategic imperative”

Problem is, your home Wi-Fi network will need to be as dependable as the cellular network run by carriers (smart Wi-Fi is designed for this purpose). Today’s home networking technology basically blows (14 year old offspring slang for "operates very poorly") for dependably carrying all this new digital multimedia content.

Not surprising, the BIG TREND that seemed to drive most of the interest at CES was the collision of IP networks and consumer electronics. CE companies clearly understand that that all the digital multimedia information going into and coming out of their boxes will be wrapped in IP. From Bang & Olufsen to Sony, Motorola to Philips, IP connectivity has taken center stage.

The problem is, these boxes don't speak IP too well. And even if they have an Ethernet jack on the back (most now come with one but don't tell you how to use it...click on picture), consumers must plug an Ethernet cable into it.  Now just try getting an Ethernet cable (or any other cable) to it.

But make no mistake, the "digital home" is real - whether you want it or not. Despite being an amorphous term made up by committee of creatives, the digital home is causing opportunities and problems for everyone in and out of the value chain.

Consumers now have a ton more services, information and products that make things better for them.  Meanwhile providers have new ways to make more money.  Everyone’s happy right?  Wrong. Today’s consumers don’t have a home network that can reliably support all this new content and all these new services. 

The pre-CES New York Times article said it best: <snip> "Getting consumers past the confusion of how to link, say, a PC to a TV will be the next big hurdle" <snip>

We agree.