The Next Decade of Bandwidth - Part IV

It won’t be long until wireless broadband has commandeered all the spectrum it can between a few hundred and a few thousand MHz. These are seen by the industry as being the most desirable frequencies. They offer high bandwidth capacity, good penetration into buildings and small antennas. Unfortunately, there are only so many of these channels available. Just like in real estate, when all the best locations are bought up expansion continues into whatever is available.

Going up or down the electromagnetic spectrum in search of more broadband capacity has its challenges. As you go down in frequency, up in wavelength, channels have to be proportionally larger to get the same carrying capacity. A 1 MHz channel at 10 GHz seems small. A 1 MHz channel at 10 MHz eats up a huge chunk of that band. At 1 MHz, there would only be room for a single 1 MHz channel.

One way to deal with this is to divvy up your transmissions into hundreds or thousands of smaller channels. This is the principle behind OFDM or Orthogonal Frequency Division Multiplexing. The system generates an array of carriers spaced out across the available spectrum. Each carrier transmits only a portion of the data. When those pieces are combined at the far end, you have a single continuous transmission again.

OFDM is the system used for BPL or Broadband over Power Line Internet services. It uses the shortwave frequencies, with some carriers omitted to avoid interference with other services. By using many, many small channels instead of one big one, any packet losses are small and easily replaced. It also offers a way to fit a wireless service into a band that doesn’t have big open chunks of spectrum. Just use the channels you can and combine a bunch of them to create a larger bandwidth service.

BPL may not be going anywhere because other wireline and wireless services are more cost effective competitors. WiMAX uses OFDM, as do LTE, WiFi and digital radio broadcasting. Seems like there is no reason a wireless equivalent of BPL couldn’t be deployed to use available lower frequencies that haven’t been previously considered for broadband use.

Going up the spectrum in frequency is also challenging. The higher you go, the more line of sight transmission becomes. The Ku band that broadcast satellites use from 12 to 18 GHz has lots of capacity, but not much ability to penetrate obstacles. Even tree leaves will interrupt service, as will a heavy rain. Wireless point to point transmissions in the Super and Extremely High Frequency bands is via outdoor antennas within line of sight.

That’s the state of the art today, but does it need to be so limited? Perhaps someone will come up with a way to flood areas with many very low power cells to ensure that a mobile antenna can receive a signal no matter where it travels. That same idea is being put to use now with mesh networks consisting of WiFi radios. Each WiFi hotspot communicates with the radios nearby to share traffic, so you don’t need backhaul connections to a central controller.

This idea of mesh networks with low power transmissions and short range coverage has application for even higher parts of the spectrum. Above 300 GHz, electromagnetic waves start to be called infrared light waves. Infrared light is the carrier used in fiber optic cables. You can eliminate the fiber and use infrared beams from point to point, as long as you have a direct line of sight. This is called free space optical transmission. With enough emitters and receivers connected in a mesh network, you can create a very high capacity network with decent coverage.

In Part I, Part II, Part III, and Part IV, we’ve seen how wireless broadband has a near insatiable need for bandwidth and how that may be satisfied by reassignment of desirable channels and more efficient use of underutilized frequencies up and down the electromagnetic spectrum. In the fifth and final part of this series on the next decade of bandwidth, we’ll take a look at wired connections that include both copper and fiber to see what else is in store for business bandwidth.

Is your business running short on bandwidth or simply looking to get a better deal? See what's available for your business location in the way of T-carrier and Ethernet copper, SONET and Ethernet fiber, fixed wireless and Hybrid Fiber Cable (HFC). Get instant bandwidth pricing up to 1 Gbps and fast quotes on other services.

The Next Decade of Bandwidth - Part I

The recent disintegration of the merger between AT&T and T-Mobile might seem like a victory for consumer choice versus telecom monopoly, but it was never really about that. Even if these two GSM carriers became one, there would still be the counterbalancing force of CDMA carriers Verizon and Sprint. It wasn’t about owning the space. It was about owning the bandwidth.

Is there any doubt in your mind about this? OK, consider that no sooner had the deal fallen apart than AT&T went out and bought $1.9 billion worth of spectrum from Qualcomm with the FCC’s blessing. Verizon just announced that they bought $315 million of 20 MHz AWS spectrum from Cox right after buying $3.6 billion for 122 spectrum licenses from Comcast and Time Warner Cable.

What’s with the bandwidth buying frenzy? It’s all about mobility. There are two laws of physics that must be dealt with to satisfy our appetite for everything on the go. First, you have to use wireless spectrum. You can’t be trailing a wire everywhere you go, even if it is a nice fiber optic cable. Second, the more bandwidth you want to support the applications, the more bandwidth you need to transmit through the air.

Not all bandwidth is created equal. We’ve gotten used to cell phone services delivered on the lower microwave frequencies around 2 GHz. The next deployments will be over the old TV frequencies around 700 MHz. That’s significant because the higher up the spectrum you go in MHz and GHz, the less those transmissions can penetrate into buildings. Light is really nothing more than a really high frequency transmission in the same electromagnetic spectrum. Just about anything will stop a light beam. At the other end of the spectrum are VLF (Very Low Frequencies) around 100 KHz that go around the Earth, hugging the surface. Those are used to synchronize our “atomic” clocks with national standards in Colorado.

The other thing you should know about the spectrum is that if you want more Mbps you need more MHz. If you want Gbps of Internet bandwidth for your mobile device, you need GHz of spectrum to carry it. This is why VLF isn’t much good for anything but sending low bandwidth voice and data long distances. Microwaves have the capacity to transmit massive amounts of data, but they tend to be line of sight. Even heavy rain or tree foliage will stop satellite TV in its tracks.

This is where life gets tough for the FCC. AM radio stations are flickering off and going dark one by one, but those frequencies aren’t of much use for video transmission. Shortwave radio is being usurped by the Internet. Only problem is that the shortwave channels don’t have massive data carrying capacity, either. What’s really desirable are those juicy frequencies between 200 Mbps and 2 Gbps. That range includes UHF television and GPS satellite bands. Note the big squabble between LightSquared’s 4G broadband and the L band GPS satellite systems. Both services want the same band around 1500 MHz for different purposes and they can’t both be there without interference.

What AT&T bought from Qualcomm was UHF channels 55 and 56, formerly assigned to broadcast television and repurposed for Qualcomm’s FLO TV. It wasn’t hard to see this coming. FLO TV was an idea that came too late. Who wants to buy a separate device that only picks up a few video channels now that you have a smartphone or tablet that can get anything over the Internet? AT&T will dump the FLO technology and press those channels into service for LTE, the 4G mobile broadband standard.

Mobile Internet is becoming a black hole for spectrum and 4G is the standard that will try to feed it. Nothing else is fast enough, although WiMAX and EVDO could be if they had the momentum behind them. It appears that nearly all carriers, including WiMAX pioneer CLEAR, are headed to the LTE standard within the next few years. Everything else will shutdown from lack of available transmission channels.

What everybody wants is a single big pipe from the Internet to whatever they are using. That could be a smartphone, a laptop or tablet computer, an Internet-powered radio, weather radar and information system in their car, high definition television set or applications still to come. Giving them that without turning over the airwaves completely to the Internet will be the challenge.

In Part II, we’ll take a look at what is likely to happen next in the battle for bandwidth, especially in the lucrative UHF TV spectrum being jealously eyed by cellular carriers for 4G expansion.

Is your business running short on bandwidth or simply looking to get a better deal? See what's available for your business location in the way of T-carrier and Ethernet copper, SONET and Ethernet fiber, fixed wireless and Hybrid Fiber Cable (HFC). Get instant bandwidth pricing up to 1 Gbps and fast quotes on other services.

FCC Confiscates Remaining TV Channels For Twitter

Shocked and horrified by the geometrically expanding demand for wireless broadband services, the Federal Commfiscation Commission exercised its powers of eminent domain today and confiscated the remaining broadcast television channels not already sold at auction. Most surprising of all was the declaration that this rare and important spectrum is going to be dedicated exclusively to the text messaging service, Twitter.

FCC Chairman Lowon B. Andwidth explained that the Commission just completed an intensive study of Internet usage and found that the Twitter service is growing at such an astonishing rate that it will soon devour all available bandwidth. “While it is bad enough that Twitter will bring web browsing, video streaming, text messaging and email to a halt,” said Mr. Andwidth, “a far worse situation will occur when Twitter users have to take a number and get in line to post their tweets. I don’t want to be in Washington when that mob of anti-social networkers descends on Capitol Hill.”

In shortest supply is wireless bandwidth. Apparently, the lion’s share of cellular phone activity involves posting messages to Twitter rather than the growth previously expected from mobile video and web browsing. One young smartphone user was heard to exclaim, “Does anyone actually surf the Web anymore? Who has the patience to read blogs and web pages? Anything worth saying can be said in 140 characters or less.”

Twitter itself issued a 130 character statement expressing delight at the government’s pre-emptive action to prevent a massive Fail Whale epidemic from sweeping the nation, and requesting re-tweets. They vowed to expand server capacity to match the new wireless bandwidth capacity as best they can. TV stations that follow Twitter have already been tweeted to cease broadcasting immediately and return their licenses. The remaining stations will be getting old school telegrams. “It’s unlikely anyone will be inconvenienced by this change,” explained Chairman Andwidth. “Everybody is on cable and satellite by now, anyway.”

Expectations are that a swift coordinated move by the public and private sectors can ensure adequate messaging capacity at least until after the fall elections. After that, the FCC will likely look at shutting down other less important uses of the electromagnetic spectrum including most AM and FM music stations. “We may even have to shut down microwave ovens and garage door openers if we can’t keep up with texting bandwidth demand,” exclaimed an obviously frustrated chairman.

Also making quick moves to protect its interests, Google announced that it is going to take over non-wireless Internet service by offering 1 Gbps fiber optic connections to every city that performed ridiculous stunts to try to win Google broadband service for its residents. “All other cities that didn’t make fools of themselves will be getting 10 Gbps Google Internet access free of charge,” according to an unofficial Google statement. It went on to say: “There is only one acceptable day for any foolishness and that is April 1.”

The Fiber Gold Rush Is On

We’ve known for some time that ultimately we’d need to move from copper wires to fiber optic strands to get the bandwidth we’ll need as technology advances. Internet service providers, Cable TV companies, medical centers, telephone switching offices and major corporations have gradually embraced this move over the last couple of decades. New services, such as Metro Ethernet, have made fiber optic connections affordable for even mid-size companies. Small businesses and consumers are now getting their turn.

Indeed, there has been one company that saw the light, so to speak, on fiber to the home and made the commitment to build out the infrastructure. That’s Verizon. Their FiOS passive optical fiber service has been moving into one community after another as a replacement for DSL, Satellite, and Cable. You can get bundles of phone, Internet and TV. Or, you can just get really fast Internet service at up to 50 Mbps download and 20 Mbps upload.

Did I say really fast internet service? Most people are getting somewhere between 3 and 10 Mbps broadband via DSL, Cable or wireless. But DOCSIS 3.0 is being rolled out on Comcast and other cable systems with 50 Mbps service offerings to rival fiber. The next tier will be 100 Mbps for those who can and will pay for it. DSL over copper telephone wire is going to struggle as the bandwidth wars speed past 100 Mbps. But there’s a new service on the horizon that will blow right past all of these broadband choices. It’s a 1 Gbps fiber optic service about to be launched as a pilot program by none other than Google.

Yes, some lucky consumers in the test markets are going to get their hands on the kind of bandwidth that makes IT managers green with envy. A Gigabit per second. That’s 1000 Mbps. There’s nothing over coaxial or twisted pair copper and certainly nothing wireless that can even come close to delivering that kind of consumer bandwidth. Even Verizon may have to change out a lot of terminal equipment to pump a Gbps into home networks.

Google swears that they are not planning to become a nationwide ISP. Their intention is to create an environment that can showcase the next developments in technology. What sort of cloud computing, virtual reality, telepresence, or HD 3D video applications need that kind of bandwidth? They’re yet to appear. But this does get past the chicken and egg conundrum, where application development stagnates because of bandwidth limitations and only cutting edge aficionados pony up for the top tiers in broadband service. Google plans to make their service cheap and plentiful so that even the common user can have a crack at this massive bandwidth connectivity.

ISP or not, it’s almost a sure bet that Google will prime the pump with some of their own cutting edge experimental apps. There’s probably all sorts of cool things brewing in the lab that just aren’t practical to launch on today’s broadband networks. Once the bottle’s open and the genies start flying out, you can bet that the development communities will fall all over themselves to get their own offerings out for evaluation.

This pitiful era of a bandwidth starved United States is about to end. That’s exceptionally good news for anyone living out in the boonies who’s stuck with dial-up, or paying a goodly price for satellite broadband and swearing at the latency. Google is, perhaps, the last straw in forcing universal fiber optic connections to the premises. Right now, smart telephone and cable companies are grabbing federal stimulus financing to trench fiber everywhere it isn’t already. It’s more than just serving the populations in low density areas who need and deserve better broadband. Another driving factor is the burgeoning wireless bandwidth demand to support smartphones, such as the iPhone and Droid. The cellular sites are bandwidth limited by the copper lines they have for backhaul. They are going to need fiber to support LTE and speed-ups in WiMAX.

How will wireless fare in the long run? It’s hard to tell how fast will be fast enough for wireless broadband. We’re not there yet. Right now, the relatively small size of cellphone screens takes some pressure off the need for massive bandwidth, even for HD video. But what about that iPad? It’s surely just the first in a whole class of tablet size products with bigger screens and more processing power. That’s going to lead to higher bandwidth demands to support content to feed the tablets. The FCC is already eyeing underutilized TV channels and other portions of the spectrum to press into service for wireless broadband. There will likely be some technology advances needed in the wireless arena to satisfy the ultimate need. After all, you can put more fiber bandwidth in the ground that there are Gbps in the entire regulated electromagnetic spectrum.

I'm Sorry, Dave. I Can't Let You Use That Channel

After the high dollar feeding frenzy of the last spectrum auction, you might have the impression that all the good frequencies are spoken for. From DC to light, there isn't a spare kHz to squeeze in one more carrier. Or is there?

If you call the FCC office and tell them you want to start a new radio-based service and then ask what band of frequencies they can give you exclusive use of, they'll probably be polite enough not to laugh in your face. Unassigned swaths of spectrum are as rare as unclaimed swaths of land. But that doesn't mean that occupied communications channels are packed cheek to jowl up and down the electromagnetic spectrum. In fact, at any given time in any given place there is plenty of room to communicate.

The problem with making more efficient use of the various RF (Radio Frequency) bands lies in the way that transmitters and receivers operate. A transmitter sends out a signal on one or more channels, which is then picked up by a receiver listening on the same channel or channels. If two transmitters try to occupy exactly the same space at the same time, interference occurs and any listening receivers hear gibberish.

You may have experienced this using walkie-talkies, or hearing two broadcast radio stations cutting in and out on the same channel. WiFi networks and wireless phones using the same 2.4 GHz spectrum can also interfere to the point where one or both won't function properly.

Some services, such as radio and TV stations, transmit continuously 24/7. Those channels truly are spoken for. But other communications and data services only transmit some of the time. When nobody is transmitting, those channels might be used by another service. Even TV and Radio station channels aren't used to the fullest. Not every channel has a listenable signal in each and every location.

The major obstacle in the way of letting more services use the currently assigned spectrum is who's going to manage the traffic? If you just let everybody do what they want on any channel they want, you'll soon have little more than a bar room brawl where the commotion is so loud that nobody can hear anything. It's channels assigned and policed by a central authority that keeps things civil.

The way cellular telephony works gives a hint at one way to let many, many users use only a few channels without interference. A central switching system tells each phone what channel to use. When that call is done or the user moves out of range of a cellular base station, the channel is reassigned to another caller.

So is the answer to establish a central switching system for all communications and have it assign channels to whomever wishes to transmit at a given time? Very Orwellian or perhaps something reminiscent of HAL in 2001: A Space Odyssey, but not very practical and unbelievably expensive.

But what about building intelligence into the equipment that will use the spectrum? The rule could be that you can use any channel allocated for a particular purpose, but you have to listen before you transmit and only transmit if the channel is clear. In deference to established licensees, if the assigned user wants the channel, you have to get off immediately.

This is the idea behind proposed "white space" systems that would squat on unused TV channels to deliver services such as rural broadband. But it can be taken further and generalized into a concept known as cognitive radio. These would be smart radios based primarily on software that would be programmed to take a number of factors into account to avoid interfering with licensed or even unlicensed users of a communications channel. As long as there is some open spectrum, cognitive radios could find it and put it to good use.

Cognitive radio is a far cry from the crystal sets that pioneered radio broadcasting. Even the latest in spread spectrum systems that use sophisticated technology fall far short of radios that could be considered deep thinkers. White space transmitters still need to prove they can work reliably before anybody is going to unleash them on even TV channels where interference is unlikely. Expanding cognitive radio technology beyond that well-defined and limited application will be done slowly and carefully.

The sophistication of low cost processing chips may well be at the point where every radio can have a brain that is so agile and polite that shared usage of scarce spectrum is now practical. What's hard is accepting the idea that nothing is going to go wrong - go wrong - go wrong.