Communications Convergence: Outstripping Wires
There was a day when telephone companies provided a dial tone, cable companies offered television stations and specialty channels, and Internet companies offered service over telephone line-based modems. Recent events make it clear that those days are long, long over.
You know, of course, that many different companies provide high-speed Internet access over cable lines, phone wire, and radio frequencies (Wi-Fi and many other standards). You may know that telcos are offering cable TV-like services in many parts of the world and are jumping through regulatory hoops to do so widely in the United States using very-high-speed DSL or fiber-to-the-home (FTTH). And you might even know that cable companies can sell you phone services in some parts of the U.S.
What’s probably unclear is how quickly all this will change.
Cable Firms Go for Voice — Several major cable operators (called MSOs for "multiple systems operators") recently penned a deal with Sprint Nextel, the merged number-three cellular operator in the U.S., to resell cell service to their customers. Any time you can put more services on a single bill, you cut as much as $20 in monthly service costs for maintaining a separate billing account. It’s easy for companies to find synergies that work because of that.
But it’s not just a single bill that’s in play. The cable firms will license TV programs they own to the Sprint PCS division to stream over third-generation (3G) cell networks to new cell phones on which you can watch programming on demand.
And it goes further: Sprint (among other cell companies) will likely start offering handsets that have Wi-Fi and cell standards built in to provide what’s known as unlicensed mobile access (UMA), a form of voice over IP and Internet telephony. With UMA, instead of a cell phone hooking up with a nearby cell tower, it senses a local (typically, an in-home) Wi-Fi network and connects, using a bit of the Internet to then transmit calls to the cellular operator’s gateway and off into the phone system.
UMA can offer better-quality indoor calls, still a plaguing problem for cell service, and enable operators to offer huge piles of minutes for calls placed using UMA, which in turn can preserve users who might otherwise switch to Internet telephony at home via Vonage or another provider. In Europe, some existing cell systems sense when a customer is using their home network versus another Internet network, and pulls minutes from a home pool instead of a roaming pool; this might also be the case with UMA, to help a UMA-based plan replace a wired phone line without increasing cost for calls made in the house.
The Broadband Wireless Picture — But wait, there’s more to wireless than just that! Cable giant Comcast recently invested in BelAir Networks via its capital development arm. BelAir makes outdoor wireless broadband equipment used to build metropolitan-scale networks for public and governmental access. BelAir announced the investment the same day that it revealed its latest products: wireless mesh access points that can be plugged directly into cable wiring and use the power that already traverses cable lines.
With BelAir gear, a cable company could add a Wi-Fi network to an entire city by connecting wireless access points into existing cable lines up on telephone poles. There are a lot of "ifs" about this: in the U.S., cable operators are governed by thousands of local franchise boards which tax and constrain the operators with specific requirements in return for rights of way on roads and poles. Some franchise agreements may allow adding Wi-Fi access points, some may restrict this, others are likely silent about it.
Many cities are already far along in their plans to have private firms build municipal-wide Wi-Fi networks, however, and cable companies may want to use their existing relationships and this new technology to offer these new networks instead of allowing a third player – after telcos and cable firms – to enter the local broadband market. Current municipal-scale networks will likely promise only about 1 Mbps each way, somewhat less the typical normal downstream speed of DSL and cable, but price the service at about the cost of dial-up today – $15 to $25 per month. (1 Mbps is from 30 percent to 300 percent higher than the typical upstream speed, incidentally.)
Metropolitan-scale networks will likely employ some or a lot of mesh networking, in which Wi-Fi access points aren’t individually connected to some form of backhaul to a central network. Instead, typically several access points are tuned to the same channel and serve both as conduits for individual users and for data to pass among each other. One of the access points is plugged into backhaul that carries data to and from the network. The disadvantage of most forms of mesh is that every hop across the mesh network until it hits backhaul repeats the same data. If user A connects to access point 1 which connects to access point 2 which connects to access point 3 which connects to the backhaul, every chunk of data from user A takes up air space for the entire cluster of mesh nodes three times. This is why mesh networks are typically used to extend a network and for redundancy and failover (when a node fails, access isn’t cut) but can’t span huge areas.
Two and a Half Billion Vibrations per Second Can’t Come Cheap – Here’s where Sprint Nextel comes into the picture again: the two companies didn’t just merge customers and operations, they merged their spectrum portfolio. The two firms controlled licenses for the 2.5 gigahertz (GHz) frequency band that covers 80 percent of the country. This band, with a starting frequency just above the tail end of the unlicensed band containing Wi-Fi and Bluetooth, was originally licensed for educational institutions and distance learning. It’s a large swath of beautiful and mostly unused space. (Licensed frequencies are reserved to the license holders to use; unlicensed frequencies can be used by anyone but only with equipment that’s passed certification by the FCC in most cases. Wi-Fi gear has been certified, but can be used by anyone, anywhere in the U.S.)
Several years ago, Congress allowed the academic and non-profit entities that controlled the regionally allocated frequencies to sublicense to commercial firms in the hopes of jump-starting more advanced telecommunications service. But many telecom firms were uninterested, and the licenses were quickly snapped up by Sprint PCS and WorldCom, with BellSouth and a fourth firm being lesser players. (All four companies together owned 90 percent of the licenses.)
Sprint and WorldCom nearly merged in 2000 partly to pool what were seen as valuable licenses. Nextel bought WorldCom’s 2.5 GHz licenses out of bankruptcy in 2003, and the Sprint Nextel merger was partly seen as a way to consolidate two smaller cell players and partly, again, as a tool to consolidate those licenses. The 2.5 GHz band is exciting to these carriers because it allows higher power to be used than is allowed in Wi-Fi, thus increasing range, and interference is impossible because the carriers own all use of selected frequencies in regions the licenses cover.
Before the merger, Sprint and Nextel, along with separately held Clearwire (a firm bought in 2004 by cellular pioneer Craig McCaw) had been experimenting with broadband wireless over 2.5 GHz in small markets around the U.S. Clearwire has started rolling out low-broadband-speed service in places like my hometown of Eugene, OR, and internationally in cities like Dublin, Ireland – areas with little broadband choice and small service areas from incumbents, but a good demographic to pay for their service.
Confusingly, the 2.5 GHz band is in the middle of a multi-year set of spectrum reform negotiations among the FCC, incumbent institutional holders who actually broadcast educational programming on it, sublicense holders like Sprint Nextel, and other interested companies. The 2.5 GHz band is inefficiently organized for the digital era, being a vestige of analog broadcasting and early data services. The new proposal would preserve some existing licenses by moving them around, but open up much more usage by other parties. This band might wind up being critical for the deployment of WiMax, a broadband point-to-multi-point wireless standard that’s just starting to move into the market.
WiMax is seen initially in urban areas as a replacement for leased digital lines used by businesses, known as T-1 lines which runs at 1.544 Mbps. With WiMax, a central base station at a high point can serve many receivers in an arc that can be fairly narrow. Some early pre-WiMax deployments – devices are just being certified as compliant with WiMax standards now – offer speeds higher than T-1s for much less money. Putting in two T-1s typically doubles capital and recurring costs with wired lines; putting in a broadband wireless connect of 3 Mbps each way might cost just 10 to 30 percent more each month than a single T-1 with less installation complexity, less capital outlay for hardware, and a quick install. In rural areas, WiMax may be used for basic broadband where a wired infrastructure doesn’t exist.
Your Television Is Ringing — You’re probably holding your head, thinking, "I just want to make phone calls and surf the Web!" Don’t worry. You’ll be able to, just in more ways, with potentially fewer bills, than ever before. Every time a set of companies promises that convergence will reduce costs, you start laughing, right? But this time, the number of different kinds of firms involved in competing with each other for your business might actually improve service and reduce overall costs.
For instance, my wife and I have shaved our combined local and long distance phone and cell phone plans over the last two years from about $300 per month (that includes all my business calling) down to about $160, while adding unlimited calling within the U.S. and to 22 countries at home, and a pool of minutes on our cell plans that we rarely exceed (and use rollover minutes from other months to avoid overages). To accomplish this, we switched long distance from per minute to an unlimited flat rate voice over IP calling plan, moved our cell phones from Verizon and AT&T Wireless to Cingular because of its rollover minutes feature (unused minutes are banked for up to 12 months), and got on the same plan to reduce the cell cost, which also means minutes used to each other aren’t counted.
The coming convergence will be weird, confusing, and overwhelming, but it’s likely to mean that most people in the U.S. and many people worldwide will see much higher downstream speeds for Internet access without increased costs – we’ve seen some of that already – and with cell calls and long distance all coalescing into one flat monthly rate substantially below what moderate users pay today. And that’s a good thing.