The quick conversion of 2G and 3G networks could be key to LTE’s success as the carrier’s chosen technology of the future.
Long Term Evolution (LTE) is the first wireless technology since cellular phones were invented that will be adopted on a worldwide basis. Even in the early days of cellular, there were two or more flavors of analog in use around the world. LTE will continue to be advanced by 3GPP and others, and we are already seeing that LTE Advanced will be in the market in the next few years.
LTE is the most spectrally efficient wireless technology to be developed and it continues to be enhanced; it offers better capacity per cell sector and higher data rates than we have ever seen before. These are the reasons LTE is being adopted worldwide as the fourth-generation standard.
Over time, not only will we have new LTE networks being built at a fast pace, we will also see that our existing 2G and 3G networks will be turned off with LTE being used on these portions of the spectrum as well. AT&T has already stated that it plans to shift its 2G network spectrum to LTE by 2015 and has strongly suggested that anyone who still has a 2G-only phone would be best served by upgrading to a phone with 3G and even 4G (LTE) technologies. All signs point to LTE becoming the pervasive worldwide wireless standard, and since it is an end-to-end IP technology, there is a belief that it will be easier to deploy, easier to manage, and easier to integrate into the wired IP network known as the Internet.
But having LTE as a global standard does not mean we will all be able to purchase a single device—a smartphone, tablet, or other device—and be able to use it anywhere there is an LTE network. In spite of the fact that we have a recognized standard in LTE, the different portions of spectrum it runs on in various parts of the world precludes the industry from developing a single LTE device that will work everywhere. The Wireless Intelligence service of the GSMA says that by 2015, LTE will be deployed on 38 different portions of the spectrum. My count, as of today, is 42 spectrum segments. These frequencies range from 700 MHz in the United States and elsewhere up to and including 2.5 GHz (2500 MHz). In the future, as TV channels in the United States are pushed down below Channel 32 (today there are 51), the frequencies in use will move down to 500 MHz and could be extended into the 3.5 GHz (3500 MHz) range.
Further, there are two flavors of LTE. One is the world-accepted cellular standard of Frequency Division Duplex (FDD), and the newer Time Division Duplex (TDD). FDD is used by most cellular operators today. FDD requires each network operator to have spectrum in two different portions of the band, one for transmitting from cell sites and the other for the user devices to transmit back to the cell sites. TDD, on the other hand, uses the same block of spectrum for transmitting to and from devices and the signals are timed to permit both sides to transmit at what appears to be simultaneously. Both technologies work and have been proven. Clearwire, China Telecom, and others are in the process of deploying TDD systems and there is some TDD spectrum available in other parts of the world, so I suspect once TDD is proven efficient, it will be deployed in other areas of the world. In some cases, as is the case with Clearwire, TDD LTE will replace existing WiMAX systems.
So we have 38 or 42 different portions of the spectrum, two different ways to deploy LTE, and small handsets and tablets. The old adage that you cannot fit ten pounds of stuff in a two-pound container certainly fits this situation.
Further, the devices will have to be backward compatible for a number of years since it will take LTE some time to achieve the worldwide footprint currently enjoyed by second-generation networks. Typically, new networks are rolled out to cover major metro areas first and then, over time, find their way into the suburbs and rural areas. In the United States we are fortunate that both Verizon and AT&T have chosen to roll out LTE faster than any other technology they have ever deployed, but we will still end up falling back to their 3G and even 2G systems as we move around and, frankly, as demand for 4G in congested areas forces some of us to the other networks.
Let’s turn now to the devices. Starting with the radios that must be in every device—GPS, Wi-Fi, and Bluetooth—next we add, for the United States, 700 MHz (LTE), 850 MHz (2G, 3G), 1900 MHz mostly 3G, and AWS-1 (1700/2100), which is fast becoming a mix of 3G and LTE. So far, we have one radio that is receive-only (GPS) and five radios that are transmit and receive. This is doable, with a caveat! Today’s technology does not permit all of the above AND both AT&T and Verizon 700 MHz LTE in the same device. Why? This brings us to the black magic part of this on which the best radio and chip engineers in the world are working. The AT&T system on 700 MHz is in the lower portion of the 700 MHz band and the Verizon system is in the upper portion of the band. While it is possible to have a chipset that covers the entire band, today it is not possible to have a duplexer (used to combine transmit and receive on the same antenna) and the filters needed to protect the receiver from interference that cover the entire 700 MHz band. So today’s LTE devices are EITHER AT&T or Verizon, but not both.
This brings us to the problem of 38 different portions of the spectrum. Qualcomm’s Gobi chipset is very capable (and there are others) and can move between both frequencies and technologies with great agility. However, in order for the handsets to be able to work they must contain multiple sets of antennas, duplexers, and filters. Some antennas can be used for multiple bands, and there is work being done to extend this BUT the most important element of any device is the antenna system. Ever notice that Public Safety radios use external antennas? Remember back to the days (analog) when each phone sported an external antenna? This is because the antenna is the heart of the system. When we began having enough coverage, and engineers worked on internal antennas, we were able to hide the antennas inside the devices. However, the efficiency of the antenna inside a device is poor so we need really good cell coverage for it to work.
Today, the factor that will limit and continue to limit the number of radios that can be fit into a device will not be the radio chips but rather the needed filters and duplexers and other related circuitry. As more, different portions of the spectrum are added to LTE, devices will become more complex and customers will have to make sure that the devices they purchase will work where they are and where they intend to be. If you recall, when Apple announced the new iPad with LTE, the folks down under had a real problem with Apple’s 4G claim since the iPad did not include LTE on their spectrum. The new iPhone 5 will cover more portions of the spectrum but it will not be able to cover them all. That is a given.