Much of 5G technology remains a moving target in a rather literal sense. NTT Docomo in Japan recently published a paper describing how, with an assist from Keysight Technologies, it is characterizing what happens in various millimeter wave (mmWave) bands when receivers, transmitters, and different kinds of obstacles–including human bodies–are all in motion.

Signal propagation and channel behavior are understood well enough when it comes to the lower frequencies that the first 5G systems will use to provide fixed wireless broadband services. When the transmitter and receiver are nailed down, anything that comes between them is likely to be passing through. A static system with occasional and fleeting variables is relatively easy to deal with. When AT&T and Verizon start providing fixed wireless broadband service in a few months, as they’ve vowed they’ll do, they will have at least an adequate understanding of the signal environments they’ll be operating in and will have configured their cellular systems accordingly.

But when either the receiver or the transmitter is moving, or both are, and when things in the environment move, no one is absolutely certain what will happen to 5G signals. That is especially so at the higher frequencies that will be used for otherwise run-of-the-mill mobile telephony and broadband.

The inability to predict channel behavior is compounded by virtue of the fact that mmWave signals don’t penetrate objects, including vehicles, trees, or bodies. What happens with bodies is a key question; what happens when there are numerous bodies and they’re all in motion is an even bigger question.

Docomo recently investigated signal propagation at 67 GHz in urban canyons containing crowds of people. In the summary of its paper documenting the research, it notes that “the effects of shadowing and scattering of radio waves by human bodies (HBs) on propagation channels cannot be ignored.”

The carrier is only one of the many companies around the world that are characterizing the behavior of channels at the various wavelengths that will be used in 5G networks, but it is also one of the few that are publishing test results. The summary of its paper provides some tidbits of data, for example that the radar cross section of human bodies fluctuates randomly over a range of roughly 20 dB, but the paper itself, of course, goes into greater detail.

Docomo and Keysight have an ongoing agreement to conduct channel sounding studies at a variety of mmWave frequencies. Keysight said it has just provided Docomo with its 28 GHz channel sounding solution. 5G channel sounding is used to make a variety of measurements including path loss, power delay profile, reflection, and various fading profiles including Doppler shift.

Keysight Technologies' 28 GHz channel sounding apparatus. Source: Keysight Technologies

Roger Nichols, who leads Keysight’s 5G practice, told EDN that it is important to not just test channel behavior in the lab, but to test channel behavior in the real world. “What does the channel look like? How does it behave? Those measurements have to be made in the same broad bandwidth that the communications link will be so, that’s 1 to 2 GHz. Testing can be done in narrowband with an unmodulated wave, but that doesn’t tell you what happens to the channel with the broader bandwidth,” he said.

Add to that multipath considerations. Keysight said that using its channel sounding solution with wideband MIMO data capture techniques enables engineers to measure angular spread with fewer measurements and improve resolution of the multipath parameters.

The variables are extraordinary just with stationary objects. The channel might be affected differently by single-pane glass, double-pane glass, and triple-pane glass, Nichols noted.

Dynamic environments are endlessly varied and far, far more complicated. How signals interact with human bodies on a crowded street might be different from how signals behave in a crowded room, he said. Then what if the moving objects are bigger, such as cars, trucks, or buses? What happens at different speeds when the objects are moving relative to each other, as might be the case with two high-speed trains passing each other?

How signals behave will obviously have practical ramifications. For example, when communicating environmental data (maps, road conditions, traffic conditions) to autonomous vehicles, would it make more sense to transmit a steady stream of data, which is difficult to do at 50 mph, Nichols noted, or might it make more sense to perform rapid data dumps at stoplights?

Clearly there’s a lot more work to do before mobile mmWave 5G is commercially practical for the full range of applications possible. The industry is working feverishly to make sure 5G works for pedestrians using mobile services, which some companies have promised as early as the end of this year or the first half of next.

Supporting the full range of advanced applications is guaranteed to be harder to do. When asked to estimate how much more for everything, including more advanced use cases such as communicating with autonomous vehicles or communications between passing bullet trains, Nichols said the industry might need another 20 or 30 months.

Brian Santo has been writing about science and technology for over 30 years, covering cable networks, broadband, wireless, the Internet of things, T&M, semiconductors, consumer electronics, and more.

Related articles