Beyond WiMAX: Gigabit Wireless


000126aWiMAX, with its promise of super-speed wireless Internet access, is nearly here. Defined as “a standards-based technology enabling the delivery of last-mile wireless broadband access as an alternative to cable and DSL,” WiMAX is capable of data speeds around 100 megabits per second, and can send that signal over several miles — though speed drops off as the range increases.

Yet research is already underway into WiMAX’s successor. Some of the most promising results so far use millimeter-wave technology, as such systems should be able to deliver data over the air not at megabits-per-second rates, but rather at many gigabits per second. That’s approaching the speeds you can get over fiber-optic connections.

Like other 802.xx IEEE wireless data systems, WiMAX uses centimeter-range radio waves to carry digital data. In the U.S., the frequencies allocated to WiMAX mean it will operate around 3GHz, giving a carrier wavelength of 10cm. This is convenient for antenna design, and practical data rates over WiMAX are in the tens-of-megabit range. ISPs are planning to use the system to complete the “last-mile” connection between their service and consumers’ homes, since it brings massive infrastructure advantages alongside speed boosts.

In contrast, millimeter-wave technology operates at radio frequencies of between 60 and 100 GHz, giving the waves lengths of 3-5mm. It’s a part of the radio spectrum commonly used for radio astronomy and high-resolution radar systems, and it’s largely unregulated and unused — particularly compared to the crowded, longer wavelength end of the spectrum.

It’s interesting for data communications for one main reason: Millimeter-wave transmissions, since they use a higher frequency than other wireless standards, fit more data into their signals at a higher rate. This is simply due to bandwidth. A higher bandwidth signal can carry more data, and chopping up MHz radio-frequencies creates signal bands with a particular width. Doing the same with GHz radio frequencies creates similar bands with a greater width, essentially creating a fatter radio “tube” through which data can be sent.

Despite obvious advantages, the difficulties of building technology that can encode signals onto millimeter-sized carrier waves means millimeter-wave communications have gone largely unused.

But that’s changing. A Columbus, Ohio-based company called Battelle has recently demonstrated a 20-gigabit-per-second data transmission system in its laboratory. It’s based on off-the-shelf technology that encodes data onto optical telecoms laser signals; when two of these lasers interfere, the resulting optical interference pattern acts a 100-gigahertz signal, in the millimeter-wave spectrum. Another company, Gigabeam, already has products available that can do point-to-point 1Gbps communications using millimeter-wave transmitters and receivers.

Millimeter-wave communications have other advantages. Antennas can be smaller than those used for longer wavelength radio waves, which is great for portable gadgets. And the signals can be more precisely directed, which even allows for a greater density of signals to occupy the same physical space without interfering.

Ultimately, why do we care about millimeter-wave wireless? Due to our thirst for more data at higher speeds. The wireless communications infrastructure we’ve been building will someday hit its inevitable maximum capacity, and data rate throttling by some ISPs shows they’re already trying to protect their wired broadband data distribution networks from being overloaded. Fiber optics can offer bigger data rates and higher capacity, but obviously don’t work for mobile solutions — something we’re also increasingly demanding as consumers.

It’s a question of numbers. The complete works of Shakespeare (around 5 MB of text) can be transmitted over typical existing Wi-Fi in seconds, while a gigabit millimeter-wave wireless network could do the same in a few tens of milliseconds. Given our need for speed, the future choice of consumers seems pretty clear.

Image of low-cost, millimeter-wave front-end module courtesy of NEC.


Jay Baker

I just moved from Portland where I had Clear wireless internet ( and they operate on a WiMAX system. I really loved how easy it was to use – all I had to do was plug a USB modem into my computer and I was online. Their WiMAX network was really fast and allowed me to be online anywhere in the greater downtown area. I don’t know much about the system that this article talks about, but I’m happy with my WiMAX connection.

Jesse Kopelman


There is a lot more going on than absorption bands when you go to millimeter wave. Remember, wave propagation is such that intensity decreases with the square of frequency. This means that going from 3GHz to 60GHz brings an inherent attenuation of 400X. To overcome this natural disadvantage you need to increase TX power — that is what Jason is talking about, when he says power-hungry. Yes, we can offset this disadvantage by substituting bandwidth (which higher frequencies have in spades) for power, but both power and bandwidth are linear factors and in the end it is very hard to overcome the exponential nature of a large frequency discrepancy. Anyway, the bottom line is that you will not be able to deploy a macrocellular topology using MMW anytime in the foreseeable future and that is what WiMAX is all about. This is the piece you are missing — topology. Mesh WiFi can already achieve similar throughput to WiMAX, the difference is that it requires 100X the base-stations. So, while it may be fair to call MMW a potentially viable competitor to WiMAX, it is just wrong to claim it is in any way a successor.

Kit Eaton

*NB the above comment is actually from me, just juggling between different accounts*

Kit Eaton

If you check out the specs, the research is going on at frequencies _between_ 60GHz and 100GHz. For sure there are absorption spectra that overlap this band: water for one, but that doesn’t cover the whole frequency range. Again, it’s mainly being investigated for the “last mile” connection between an ISP and your home– imagine an external box on you house aimed at an ISP’s antenna– that’s also why the article references “greater density of signals to occupy the same physical space.” Powerhungriness is a function of research time spent: in time power requirements will erode.


Gigaom must do better.. This guys is smoking..

– 60GHz is power hungry – Read: Not for mobile
– 60Ghz requires line of sight – even with beam forming
– 60GHz doesn’t work when it is raining – water kills the signals, very susceptible to humidity

60GHz has some future for P2P HDMI replacement but that is it.. Om, you gotto do better man.. We are spending time on your site..

Robert J Berger

It is literal line of sight. No obstructions at all, not even your body. Its not a mobile tech at all. And its pretty short distance as well unless you have well engineered static antennas. Great for interconnecting buildings in a downtown for instance. Possible rural backhaul with lots of repeaters. Nothing that will solve the US lack of broadband. We need fiber to the home and office for that. And the longer we waste time letting the Carriers and CableCos screw us, the more we will fall behind the rest of the world.


To add to the previous commenters’ posts, the propagation characteristics of a 60 GHz wave are fine for very short-distance applications or (slightly longer distance) line-of-sight applications. As a wireless replacement for a 1-meter HDMI cable? Fine. To communicate with a mobile phone that works indoors, or even inside a cardboard box? Not very likely.


Check out the way most cellular towers are set up next time you are driving. They are practically Line of Sight, to speak to the next cell tower. It would be more obvious if you were at the top of the tower, but it is perfectly plausible merely driving by the things.

Jesse Kopelman

Nothing based on millimeter wave can be WiMAX’s successor. WiMAX is a point-to-multipoint technology and millimeter wave is pretty much only useful for point-to-point.

Q dub

60-100GHz? Wouldn’t it require line-of-sight? That serves an entirely different business purpose from mobile wireless…

Comments are closed.