In my last column on 4G cellular services, I talked about a few issues relating to definition and timing. The bottom line, I think, is that it doesn't matter what you call it or exactly how you define it
The WiMAX community would argue that their technology is the first that can really claim to be 4G, and I'm going to agree if we limit the discussion to wide-area wireless networks -- those that are, or compete with, cellular. But a perhaps surprising conclusion, as I noted last time, is that Wi-Fi, when deployed on a metro-area scale, is really the first 4G technology. And metro-scale Wi-Fi is going to become ubiquitous on the same timescale as 4G will.
I'm still going to argue that the combination of Wi-Fi and a 3.5G or 4G technology will be the key to mobile broadband nirvana, coupling the coverage of a large-cell, licensed bandwidth cellular or WiMAX implementation with the capacity-oriented capabilities that really define Wi-Fi. 802.11n, when deployed outdoors -- and it will soon be -- is going to provide service that is indistinguishable from landlines (cable and DSL) for most people. But Wi-Fi in any form won't be able to provide ubiquitous, seamless coverage in any location, owing to limited transmit power and other limitations of the unlicensed bands, so cellular or WiMAX will need to fill in the rest of the picture. Dual-mode handsets, with both a wide-area technology and Wi-Fi together, will constitute the majority of new sales by 2011.
But let's explore the directions for wide-area 4G in a little more detail. As I noted above, WiMAX qualifies, in my book, as 4G – all IP, broadband, mobile, with support for time-bounded services. On the cellular side, we still have two camps, CDMA2000 and UMTS. CDMA2000, embraced by Sprint and Verizon in the U.S., includes EV-DO, which provides raw throughput on the downlink of up to 3.1 Mbps in EV-DO Rev A, now being deployed.
UMTS today features the very powerful High-Speed Packet Access (HSPA), which in theory can offer raw downlink throughput of up to 14.4 Mbps. Both of these technologies dedicate a cellular channel just to data, and HSPA is much faster because UMTS channels are four times wider that CDMA2000s. We call both of these 3.5G because their raw throughput exceeds the upper bound of 2 Mbps that defines 3G, but neither is designed to replace the 3G voice services already in place.
Things start to get interesting with EV-DO Rev B and Rev C, the latter also known as ultra-mobile broadband (UMB), and with the next edition of UMTS, known as long-term evolution (LTE). Rev B gets us to 75 Mbps, LTE to 100, and UMB to an amazing 280 Mbps. And now there's talk of moving WiMAX into the Gigabit range, via the IEEE 802.16m effort! But I need to stress again that such numbers are not designed to be real to us mere mortal users -- they are peak speeds. But they are indicative of the industry's ability to move forward to astonishing levels of throughput in the interest of more capacity and better time-bounded performance -- and these are really what 4G is all about.
About the author: Craig Mathias is a principal with Farpoint Group, an advisory firm based in Ashland, Mass., specializing in wireless networking and mobile computing. The firm works with manufacturers, enterprises, carriers, government, and the financial community on all aspects of wireless and mobile. He can be reached at email@example.com.
This was first published in April 2007