Smartphones and tablets have surged in popularity as cellular carriers continue to develop their fastest 4G networks to date, contributing to a mammoth increase in cell networks’ data traffic. And there’s no end in sight. Indeed, the proliferation of even more connected devices is expected to keep straining cell networks’ data capacities over the next 10 years and beyond.
To accommodate this growth, researchers at wireless industry suppliers and leading universities have been working on the next generation of wireless network technologies beyond 4G, including new cellular network architectures and new varieties of communications chips for handset innards. It all could be deployed and begin to benefit consumers by the start of the next decade.
1,000x Data Capacity
“We went from 1G that was analog voice-centric [network technology] to 2G voice-centric with some data, to 3G with a more data-centric approach and then 4G with megabits higher data rates and a real data focus,” says Rasmus Hellberg, senior director of technical marketing at Qualcomm Inc. in San Diego. Next, “5G is about bringing the network closer to the user, by adding many small cell network access points everywhere, indoors and outdoors. That’s really the main way to get more performance, high capacity and low overhead.”
Globally, data consumption on cellular networks has doubled every year for the last couple of years and it is expected to continue growing at that pace for the next 10 years, heralding a 1,000-times increase in cellular data consumption over the next decade, Hellberg says. So Qualcomm is focusing its research on developing new ways for cellular network operators to cost-effectively boost their data capacity to meet this increasing demand—solutions that might be different than the traditional ways of increasing capacity that we have today, Hellberg says.
These new solutions center on adding small cells to the macrocells that already make up cellular networks in neighborhoods, and finding new ways to use fallow spectrum as well as finding new spectrum to tap.
Small cells include femtocells used by consumers at home to be shared with other users outside the home, as well as picocells which can be deployed outside on fixtures like lamp posts. “If we can get penetration of 20 percent of the households deploying indoor small cells and we also capture the outside part of that, [we] can get 1,000-times more capacity if we add 10-times more spectrum, compared to a macro-only deployment,” Hellberg says. “Macrocells are the foundation,” he explains, but there is a physical limit to how many can be added to a network so they must be supplemented by those neighborhood small cells of “different flavors, different capacities” in “heterogeneous networks” or “hetnets.”
But to make such hyper-dense networks operate efficiently, there also is additional technology needed to make them self-organizing and to mitigate interference between small cells, so that the user experience is good, Hellberg adds.
For spectrum, he says, these small cells will also need to access higher frequencies, notably in the 3.4GHz to 3.8GHz range and eventually in the 60GHz range using a new form of Wi-Fi known as 802.11ad.
All of this is being tested in the labs at Qualcomm. The company is planning to launch a demonstration hyper-dense network in San Diego beginning early in 2013, Hellberg says. “Probably the most visible difference between a 5G system and a 4G system will be the number of small cells [used] to deliver that extra network capacity, because using small cells gives a greater incremental boost to capacity than do alternative technology options,” concurs Boyd Bangerter, director of Intel Corp.’s wireless communications lab in Hillsboro, Ore.
“We have a set of pretty specific metrics that we’re looking at, that we’re hoping will be an industry definition of a 5G system,” says Bangerter. “In the past, definitions of technology like 3G and 4G have always been very numerical,” he says, referring to hard scientific numbers such as peak data rate, average spectral efficiency (the number of bits per second delivered in any given amount of bandwidth) and latency. “What we’d like to do for 5G is introduce some new requirements that are more user experience-oriented.”
These would include, for example: making video running over a cellular network a more enjoyable experience for more users; a more uniform connectivity experience in terms of signal quality and data rate compared with the variable quality of 3G and 4G connections; and improvements to data rates, spectral efficiency and network capacity (the number of users who can share the network at once).
One of the technologies Intel is developing to meet these goals is called a cloud RAN or radio access network, Bangerter says. Demonstrated at the Intel Developer Forum, it virtualizes a cellular base station on an Intel architecture-based server blade and allows network operators “to move base station capacity to where it’s needed,” he explains. “So if one base station is experiencing heavier traffic than another, you might be able to allocate more computing resources to that particular base station at that particular time and then back off later. It’s a more efficient use of computing resources in the cellular infrastructure.”
Network Operators Test
“Historically if you wanted to deploy lots of infrastructure among places, it’s become as much of a real estate problem as a technology problem. You have to find places to mount these things and you have to figure out how to get the backhaul,” Bangerter says. To that end, he says, a developing technology named “Massive MiMo” could be used. It employs multiple antennas at each end of a wireless link—such as on a small cell or at a base station—to steer wireless signals in narrow beams of energy toward specific directions, to reduce interference and boost network capacity.
Cellular network operators worldwide, as well as infrastructure equipment providers and university researchers are working on all of these 5G technology choices, Bangerter notes.
AT &T is one. “There’s a lot of innovation and research that’s being done in terms of the evolution of both the wireless network and the backhaul and the core that’s associated with it,” says Kris Rinne, senior vice president of network technologies at AT &T in Atlanta. “We have several trials going on right now [regarding] small cells. In a lot of ways it’s the miniaturization of the infrastructure that we already have in place.” But there is still the challenge of determining whether the user is moving at a high-speed or is stationary, and then instantly determining on which “layer of the network” to connect him—the small cell or the macro network, Rinne says. Stationary users inside and outside would be connected with the small cell. Those in motion would connect with the macro network.
AT &T also is working on both the sort of beam-forming technology described by Intel’s Bangerter and “self-optimizing” or “self-organizing” network technology. The latter enables the network to intelligently trigger a mobile device to switch between using licensed (cellular) spectrum and unlicensed spectrum (such as Wi-Fi), Rinne says.
Carrier aggregation, another technology in the works at AT &T, will also help boost network capacity, Rinne says. It brings together spectrum in different frequency bands so that they act as one contiguous band. But this development hinges in part on the deployment of a self-optimizing network, which will help determine when to utilize which frequency band to serve customers’ needs, she says.
Calling on Academics
To be sure, it’s not just the wireless industry pursuing these wireless innovations beyond 4G. Last September Intel invited select academic researchers from top universities to collaborate on these new technologies. The company issued a request for proposals on “technologies for future wireless networks and devices” that will “impact the development of beyond 4G wireless (B4G) standards and products.”
Specifically, Bangerter says, Intel asked the academics to propose technologies that would enable access to new spectrum and to improve the overall power efficiency of cellular networks.
In Intel’s estimation, it will be about 2020 before consumers toting smartphones will see the benefits from any of these 5G technology developments, Bangerter says. The earlier goal of cellular service providers will be to make a return on their investment in 4G, which is still in the process of rolling out.