Every decade or so since the first cellular networks appeared the companies that make mobile devices and the networks linking them have worked out new requirements defining transmission speeds, capacity and other technical characteristics. Each new set of requirements is referred to as the latest “generation.”
Today’s fourth-generation, or 4G, wireless digital networks made it possible for smartphones and tablets to deliver voice and data communications with bandwidths measuring many millions of bits per second. Specific data speeds vary by carrier but most networks enable users to download a file containing a full-length movie—more than one gigabyte in size—in less than 10 minutes.
The next generation—5G wireless—will have to deliver a huge leap in performance to handle surging mobile network traffic, much of which will be large multimedia files. According to Cisco Systems’ most recent Visual Networking Index (VNI), mobile data traffic will grow 10-fold globally between 2014 and 2019, reaching 24.3 exabytes per month worldwide in 2019. (An exabyte is one billion gigabytes.)
The details of 5G are a long way from being decided but it is expected to provide Internet connections 40 times faster and with at least four times more coverage worldwide than the current 4G Long Term Evolution (LTE) wireless communications standard. Even without a clear definition of 5G, testing is underway or in the works in places including Finland, Russia and South Korea.
One of the most promising potential 5G technologies under consideration is the use of high-frequency signals—in the millimeter-wave frequency band—that could allocate more bandwidth to deliver faster, higher-quality video and multimedia content. Other lines of research seek to enable a single mobile device to simultaneously connect to multiple wireless networks to boost connectivity and speed.
Mustafa Cenk Gursoy, an associate professor in Syracuse University’s College of Engineering and Computer Science, and researchers at The Ohio State University have received about $678,000 in funding from the National Science Foundation to study ways to more efficiently access the radio spectrum over the next three years.* Scientific American spoke with Gursoy about the need for 5G and the role that millimeter-wave frequencies, in particular, could play.
[An edited transcript of the interview follows.]
Why do we need a new wireless standard?
The motivation behind a new standard is the exponential growth in wireless. We’re talking about billions of users, billions of devices and billions of connections. That’s something that a new standard has to address, because 4G is not going to be efficient enough to handle this much growth, much of it due to mobile video traffic. Smartphones, tablets, social networking sites and video-sharing sites have helped mobile video traffic become more than half of all mobile traffic.
On top of this, people have really high expectations for wireless services. They want a high level of reliability, low levels of latency [delayed uploading or downloading of content] and constant connectivity—anytime, anywhere. The Internet of Things, where new types of devices are connected digitally, as well as the increasing use of mobile technology for health care, smart power grid and vehicular networking create new expectations for wireless, especially when it comes to speed and reliability.
Any move to 5G wireless technology is still years away. What should we know about the standard at this point?
There’s been an increase in interest in 5G within the past year, particularly from the research community. And telecom companies in the U.S., Europe, South Korea, China and Japan are interested in designing, testing and implementing these kinds of systems, so there’s definitely some momentum building. It’s not definite what 5G will be but people are talking about candidate technologies and what needs to be addressed in this next-generation wireless standard. And it might not be that far away—some companies are thinking they will have 5G systems up and running by 2020. That’s not really a lot of time.
How will 5G differ from 4G?
One difference will be that 5G may move wireless signals to a higher frequency band, operating at millimeter-length wavelengths between 30 and 300 gigahertz (GHz) on the radio spectrum. That’s going to open up a huge amount of bandwidth and alleviate concerns about wireless traffic congestion. Radar, satellite and some military systems use this area of the spectrum currently but it’s definitely less occupied than the spectrum currently in use. In addition, whereas 4G supports hundreds of megabits-per-second data rates, 5G is promising data rates in the gigabits-per-second range. It may not support those higher rates at all times in all places, but it will lower latency rates overall.
Are there drawbacks to wireless devices operating at such high frequencies?
Generally, as you move to higher frequencies, transmission range gets shorter—hundreds of meters rather than kilometers. And signals are unable to penetrate walls easily. Some hardware components, such as analog-to-digital converters, might also be expensive. We are still learning about millimeter wave and are testing its capabilities. Another challenge is if the transmitter and the receiver don’t have a line-of-site connection, there is a lot of attenuation [loss] in the signal. We’re conducting performance analyses to better understand the communication reliability and plan to publish a paper in the fall at the IEEE Vehicular Technology Conference in Boston.
What can be done to overcome these limitations?
There has been a trend toward small cells [portable base stations often called microcells, femtocells or picocells, depending on their ranges]. Millimeter waves can take advantage of these technologies, as they are better suited for transmission over relatively short ranges. High-frequency signals can also be reused across short distances by different cells in a network, meaning the available spectrum is used more efficiently. In addition, antenna size is inversely proportional to frequency size, so higher-frequency signals would require smaller antennas. You could pack more antennas into devices. That enables directional transmissions—you could actually steer the signal in a particular direction. This could overcome the loss of some of the signal transmission strength. More than one antenna operating in the same frequency range can also send multiple streams of data, increasing the data rate.
What research are you and your colleagues doing in the area of millimeter waves?
This first year we are learning the characteristics of communication in the millimeter range of frequencies and doing some performance analyses of millimeter-wave networks. For a short range, with devices in fixed positions, millimeter waves can connect devices to a network. The challenge is delivering this service to a user who is walking or driving. I also want to see networking scenarios where you can actually support multimedia traffic in a mobile environment using millimeter wave.
How does millimeter wave improve energy efficiency?
The use of directional transmission between the base station and a mobile device reduces signal interference, and that might account for the reduction in energy use we’re seeing. When you establish a direct link and suppress interference, you can send data at higher rates for a given transmission energy level. Therefore, throughput per unit energy increases and hence energy efficiency improves. In such an analysis, it is also important to take into account possible increases in hardware energy consumption due to operation at high frequencies. Energy efficiency is very important here as well because of the growth in the number of users and devices—and efficiency should be considered with any new standard.
*Editor's Note (6/23/15): This sentence was edited after posting to correct the amount of NSF funding. The Ohio State researchers received about $460,000 in funding while Cenk's Syracuse team received more than $218,000 in funding.