Is Broadband the Answer?

Perhaps it is time for us to reassess how we look at spectrum and look at the types of services that are needed, that are in use today, what broadband can effectively handle, and what is best left to other technologies.

Okay, Sprint is now 70% owned by SoftBank of Japan and Sprint owns Clearwire, so together SoftBank spectrum holdings are the largest in the United States at this point in time. Meanwhile, T-Mobile is busy integrating MetroPCS into the company and Leap is perhaps a takeover target for T-Mobile or Dish Networks. Dish, for its part, has not given up its quest for a terrestrial broadband network and LightSquared is still lurking somewhere on the edges of terrestrial broadband.

Meanwhile, Google is experimenting with high-altitude Wi-Fi balloons to cover rural areas and it has been reported that the cable operators are getting ready to deploy an additional 250,000 Wi-Fi hotspots in an attempt to extend its wireless footprint, yet again, and of course, FirstNet, the Nationwide Public Safety Broadband Network (NPSBN) license holder is looking at public/private partnerships to help fund the build-out of this 700-MHz broadband network.

Further, the FCC is preparing to hold an auction for some more broadband spectrum, followed by the incentive auctions that will clear a number of TV channels in the channel 30-50 range for even more spectrum to be auctioned. The FCC is also looking at opening up more of the 5-GHz spectrum for unlicensed broadband services. Finally, the NTIA and FCC are engaged in following the Executive Branch directive to “find” even more broadband spectrum, and to experiment with spectrum sharing—commercial operators making use of Federal Government spectrum on a non-interfering basis where practical.

SoftBank says it will be investing $16 billion in the Sprint/Clearwire network over the next two years, most of which will be in LTE infrastructure and network build-outs. The goal, of course, is to increase the number of subscribers Sprint currently has and put it in a position to join the big two (AT&T and Verizon) as the top network operators in the United States.

There is a mad rush to mine more spectrum for broadband services. The Executive Branch, Congress, and those presently at the FCC believe that broadband is the only spectrum with any future value and that all things wireless will make use of broadband spectrum going forward. Many outside the wireless industry—Microsoft, Google, Apple, and others—would really like to have a bigger role in wireless but so far they have danced around the edges of network operation and not made any forward progress. Some of this is to be expected. It would be fairly straightforward for any of these companies to make a deal with one of the existing network operators. However, these companies really want access to all wireless customers and not be limited to those on a single network.

Over the past few years, Google made an attempt to bid on spectrum in the 700-MHz auction, launched a Wi-Fi network in Menlo Park, was part of a group that convinced the FCC to permit unlicensed broadband in unused TV channels (white space), and now says it can control balloons in rural areas of the world to provide Wi-Fi access. It is reported that Google is actually moving forward with testing in rural areas of New Zealand. Google, of course, developed the Android operating system that now accounts for more than 50% of all smartphones sold, and while it gave away the operating system, it is making money on applications and by driving eyeballs to its site. And we should not forget that Google bought Motorola’s cellular group, but so far nothing spectacular has come out of that purchase. Google does own a lot of dark (unused) fiber around the United States and there is real value in this. Wireless, after all, is really about the last mile, ½ mile, ¼ mile, or 300 feet. The rest of the infrastructure is about fiber and microwave for backhaul. Cell sites, femtocells, and Wi-Fi access points must be connected in order to work. There has been a lot invested in mesh networks—Wi-Fi networks that use a series of hops between access points to minimize the number of connections needed—but still, at the end of the day, if wireless is not attached to something to make the connection back to the network hub and the Internet, it won’t provide connectivity.

The Rush to Broadband

With all eyes on broadband, are we setting in motion a series of events that could cause us to make a number of mistakes moving forward? One mistake has already been made. When Google, Microsoft, and others convinced the FCC to authorize TV white space broadband, a number of companies invested heavily in this technology. The idea was to put unused TV spectrum in service for faster, unlicensed broadband services. At the time this was being discussed, there were many areas of the United States where there was sufficient TV white space to build a business. However, today, TV white space technology cannot be used in the top eighty or so metro areas because there are not enough empty or unused TV channels. Outside of these areas, it seems to be working okay, but when the FCC starts relocating the TV stations from, say, channel 32 to channel 52 down below these channels, how many white space systems will be forced to shut down? Remember that when the TV white space was released, the FCC also changed the spectrum allocations for wireless mics, although it is clear that many are still operating in what is now the wrong portion of the spectrum. Users of the technology have no idea that their devices are now operating illegally.

What I am suggesting here is that everyone with a say in the direction the wireless industry is heading seems to be caught up in a feeding frenzy spelled b-r-o-a-d-b-a-n-d. Perhaps some of what is being done should be reviewed, weighed, and vetted against the types of wireless communications really needed. Do we really need to move voice communications, which uses very little spectrum, into a broadband system? Do existing wireless voice and slow-speed data users have to abandon the spectrum they are licensed to use and instead obtain a service contract with a commercial operator? Can broadband really provide all of the services that are not being handled by narrowband communications? Services that, in reality, do not need broadband capabilities?

There is another wrinkle in all of this. Not all of the spectrum licensed today is truly suitable for mobile broadband services. Once you get below about 500 MHz or so, antennas must be bigger and filters and other components are larger. Perhaps some of this spectrum could be used for point-to-point broadband, but aren’t we better served to make use of this spectrum for narrowband services that already exist and, in the case of, for example, Public Safety, are critical in nature? On the high side of the spectrum, when does it become impractical to use for mobile broadband? Is spectrum above 3500 MHz really well suited for broadband services, except perhaps for small Wi-Fi nodes? At what point do the characteristics of the spectrum mean that it is of little or no value to commercial broadband operators? I believe we should perhaps limit our quest for everything broadband for wide-area communications to between 500 MHz and 3500 MHz.

This leaves spectrum below 500 MHz for other uses including narrowband communications. Suppose, as an example, that during the TV station relocation, TV channels 2-9 were reallocated not to broadband but perhaps to narrowband communications for voice and slow-speed data. These TV channels are not well suited for HD TV in any event because of the location of this spectrum and the fact that it is more prone to ghosting and multi-path issues. This spectrum would be used for many different forms of communications. The idea is to make use of spectrum for its highest and best use. This means being realistic about the spectrum frequencies, the propagation, and how much is available as contiguous. By that I mean how much spectrum is available in a given portion of the spectrum. How much spectrum is needed for effective broadband communications? Even with spectrum aggregation techniques, combining different portions of the spectrum to achieve more bandwidth only works if there is enough bandwidth in each portion of the spectrum to be used.

When I read about the Government’s spectrum mapping and inventory efforts, I never see that the spectrum is also being reviewed for the type of use for which it is best suited, or if spectrum in that band will penetrate buildings or simply bounce around and cause interference. Those cataloging the spectrum seem to be undertaking the task in a vacuum. Perhaps it is time for us to reassess how we look at spectrum and look at the types of services that are needed, that are in use today, what broadband can effectively handle, and what is best left to other technologies. Perhaps we should ask if what we want for the United States is wireless communications where everyone who uses licensed spectrum must use commercial spectrum held by network operators and if we must then pay them for the right to use the spectrum.

As we move forward, are we in too big a rush to declare broadband as the only viable form of wireless communications going forward? Are we about to make some serious mistakes by making this type of assumption? What happens to system reliability if we are all pushed onto broadband spectrum? One thing that seems to be lost on the ‘broadband is everything’ crowd is that broadband systems are more prone to failure due to storms, earthquakes, fires, tornadoes, and man-made disasters. During Sandy, and many other storms, cell systems went down and there is currently is no fallback when using a cell phone, tablet, or wireless notebook. The typical cellular or wireless broadband network has at least eight different points of failure between the back-end of the network and the device in a customer’s hand. Narrowband voice systems, on the other hand, have several available modes of fallback for what I call graceful degradation, and the tower sites, in most cases, are capable of radio control rather than having to rely on telecom or other forms of connections. This makes narrowband systems far more resilient than typical cell and wireless broadband systems.

There is work underway in the standards bodies to implement several different forms of fallback into LTE broadband systems. However, once the standards have been approved, it will be up to each individual network operator to implement these changes to the networks and it will all come down to cost versus the benefits of adding this capability when it finally becomes available. In the meantime, cellular and wireless broadband systems, today, are designed as best effort systems while most narrowband communications systems are more robust and can generally withstand more severe weather, other acts of nature, and man-made incidents.

The bottom line is that we all need to be aware that sometimes rushing into the future without regard for what the consequences may be is not always the right course of action. Perhaps, even with the increasing demand for broadband services, we should all slow down, take a deep breath, and reassess where we are heading. More and more people rely on wireless devices—more than 35% of the U.S. population doesn’t even have a wired phone, and many do not have a wired or cable connection to the Internet. This places an additional burden on commercial wireless operators, regardless of whether they are providing nationwide, regional, or even Wi-Fi services. Let’s not be in such a hurry to turn all things wireless into commercial broadband systems. Yes, we need to find more spectrum for commercial networks, but it should not come at the cost of destroying or degrading existing capabilities that already function well. Let’s strike a balance between broadband and reliability. Let’s think about what we are rushing to achieve and perhaps slow down a little, regroup, and rethink our wireless future, lest there be a possibility that when we need it most it will fail us.

Andrew M. Seybold