What can 5G do for me?

This is part two of a two-part series that I am writing on the subject of 5G (or fifth-generation wireless technologies).

In Part 1, which I wrote a couple of months ago, I provided a broad-brush overview and a historical perspective of the five generations of wireless technologies that we have experienced so far, and what was new and different in the latest fifth-generation all from a layman's perspective. It is a good primer for understanding 5G, but not a required read to understand this article.

As a bridge for those who have read part 1, I explained how the International Telecommunication Union's specifications, with the help of a spider chart (reproduced in Figure 1 below) compares IMT-2020 (commonly called 5G and colored with the darker blue shade) with IMT-Advanced (commonly called 4G and represented with the lighter blues in the chart) as a way to understand how much 5G has improved over 4G along eight different axes.

Figure 1

To follow along with this article, it is not necessary to understand what the specific axes mean, or what they do from a technical perspective. Suffice it to say, that 5G makes 4G better in every conceivable direction by an order of magnitude!

Now let's peel the onion a little bit more. First, 5G categorizes the features that are enabled with these technologies into 3 logical groups, or technology islands namely:

• Enhanced Mobile BroadBand (or eMBB)

• Ultra Reliable Low-Latency Communications (or uRLLC)

• Massive Machine Type Communications (or mMTC)

In Figure 2 below, I have color-coded the axes from Figure 1 above, for illustrative purposes (yellow, green, and blue), that showcase which technologies are prominent and associated with the three groups above. The reason this pairing of technology axes is important is that they are expected to combine together to provide new technology platforms on which to build the next generation set of applications and services.

Figure 2

Now let us explore each of these application areas.

Enhanced Mobile Broadband (or eMBB)

Think of this group as the natural evolution of the previous cellular generations: meaning bigger, faster, better. In other words faster download speeds for your mobile experience than previous generations. I have color-coded the axes that primarily contribute to this group in yellow in figure 2. That is not to say that the other axes don't contribute to this group, but rather that the yellow identified axes are the primary ones that enhance this group. The single handle that will help you remember this group is High Data Bandwidth Rates.

The flagship example of the features in this group, as a layman can relate to, is as follows: Imagine you are racing across the countryside in a high-speed rail like the mini-Shinkansen at 120 km/h. Imagine also that you are watching a 4K Ultra-Hi-Def movie (approximately 14Gb of data) on your cellular device while you are cruising. This is possible today if you have pre-loaded the movie onto your device while in the comfort of a high-speed Internet connection at home, which would have taken about a half-hour using 4G technology or 15 minutes using a wired internet connection. Now also imagine that the train passes through a station at full speed (let's say it takes about 6 secs); If the station was equipped with the 5G introduced millimeter wave transmitters along with multibeam array antennas and the transmitter tower in the station is within a few meters of your train, and your phone has the corresponding multi-beam 5G capable antennas, between the time the train enters and leaves the station, you could have downloaded an entirely new movie into your phone!

And if you think this is likely to be way out in the future, think again! Although a scaled-down version, NEC demonstrated this exact scenario back in December 2018!

The feasibility was first demonstrated by Mitsubishi Electric and NTT-Docomo back in November 2018, of a 27Gb/s maximum data throughput at a distance of 10 meters using the 28 GHz radio frequency with 16-beam spatial-multiplexing antennas!

But remember, you can't get these download speeds over any distance over 10 meters.

Figure 3

Ultra-Reliable Low Latency Communications (or uRLLC)

The features that are enabled by this group are brand new to cellular systems and we haven't even fully understood all of the ways these features will be deployed. The axes that contribute to this group are colored in green in figure 2 above.

As the name implies, the features that best describe this group are extremely Low Latency, highly secure, and reliable communications. While these terms are subjective and hard to relate to, the following graphic in Figure 4 below by the GSMA alliance, gives you a sense of the latencies that are required by different applications. In this graphic, the Y-axis shows the latency (called Delay in the graphic), while the X-axis shows the Bandwidth. The example applications in the horizontal band between 10 milliseconds up to 1 millisecond are examples of what low latency can do for you. You will see then that none of these applications exist today, but companies are furiously working towards building these applications and services.

Figure 4

One of the better examples of what is feasible is demonstrated in this really cool one-and-a-half minute video on the left by Ericsson at the Mobile World Congress in 2017 explaining a Tele-Health application where a specialist doctor could examine patients remotely with the use of Virtual Reality Glasses, a tactile enabled remote glove, and an extremely low-latency network connection, that would be available only with 5G!

Another example that I found interesting was also a demonstration shown in the video on the right by Ericsson where they had a remote-controlled car that was being operated from 50 km away. Video streams from the race track where the car was physically located and sensors and actuators on that car were fed into a haptic feedback system and automobile controls to the remote operator. I found it fascinating that the total latency in the system they demonstrated is about 50 milliseconds of which about 40 milliseconds is the time required to computationally compress and decompress the video streams. The network itself is transmitting at about 3-4 milliseconds!

But to really understand the context of latency and human interaction, I found this article that Ericsson shared on its website fascinating. It says that the typical human response latency is about 200-250 milliseconds. They even have a site where you can test your own response latency. Mine was about 280 milliseconds. That article further says that using gaming as an example, an MIT study identified that the human brain processes and recognizes latencies of up to 15 milliseconds as instantaneous. As the latency drops to 75-100 milliseconds, the human brain will consciously perceive the lag.

In summary, the human understands latencies of 15 milliseconds as instantaneous; The new 5G capable network will now be capable of stream data remotely at a 1-millisecond lag, leaving about 14 milliseconds for applications and other sensory networks (tactile, visionary, audio, etc) to add value and improve services. I suspect that this will be where a lot of new and unprecedented innovation will occur in the next 5 - 10 years.

Massive Machine Type Communications (or mMTC)

This third and final set of technology features are axes that are colored blue in Figure 2. This is also sometimes called enhanced Machine Type Communications (eMTC); The easiest way to think about this technology is as the next generation of cellular Internet-of-Things. Internet of things has been around for over a decade now and seems ubiquitous. Every single business has an angle to this. So what is new and different? The simple answer is scale. One of the biggest impediments to the growth of the Internet of things is the demand for power namely:

• To power to the sensor and the actuator on the endpoint devices

• To power the transmitter and the receiver

• To power the intermediate aggregator points

In order for these objects to be remote and mobile (think a bell on a cow), they would have to be battery operated and if batteries have to be changed every 6-8 hours, that is not a very scalable solution. Many new transmission technologies, called Low-Power Wide-Area-Network (LP-WAN) solutions have been proposed over the last 10 years (LoRa and SigFox are examples) and deployed at a tremendous cost but are not interoperable with other wireless deployments or require custom hardware solutions. The primary reason why these solutions had to be conceived and deployed was that the then-current cellular technologies (like 3G, and 4G) were extremely power-hungry and you need a power plant to run these cellular base stations! (okay that is an exaggeration, but you get my point)

But now with the introduction of mMTC in 5G, this changes the game! You can actually have extremely Low Power devices and can scale to hundreds to thousands of devices transmitting and receiving at extremely low power! While the actual cellular radio protocols related to low power were actually introduced with 4G itself (called NB-IOT or NarrowBand-Internet-of-Things, and LTE-M or Long-Term-Evolution-for-Machines), they are most commonly referenced in this mMTC context. Batteries now could last months and maybe even years and yet transmit/receive using the same cellular infrastructure that is already deployed!

Another aspect of scale is the number of devices. According to the 5G specifications, the networks must be able to support 1 million devices within one square kilometer!

While the providers of these technologies see tremendous value and applicability in Smart Cities (think millions of sensors across the city, like maybe every traffic pole, or streetlight, or parking meter), Smart Agriculture (fitting every cattle with a sensor, or measuring the moisture in every acre to determine the right amount of watering), or even Smart Energy, or Smart Manufacturing (Industry term is Industry 4.0), I will highlight one that I found fascinating.

Here is one example of what these ubiquitous sensors could provide!

In this video on the right, in 2016, researchers from the University of California at Berkeley demonstrated a smart neural dust sensor, which is the size of a grain of sand, that has been implanted inside a nerve. Responding to ultrasound signals sent to it, the sensor records and returns electrical signals in the nerve which are then analyzed to understand muscle movement. The thought is that one day, paraplegics can help move their robotic limbs using muscles with the aid of these implanted sensors. Imagine further that one could do this to help patients with bladder control or heart function disorder!

While this may all seem far into the future, or exclusively in the realm of research, I was surprised to learn that there is actually a company today, Bionaut Labs that is building an implantable robot to deliver drugs to specific parts of the human body! Take a look at the video on the left that talks about the science behind their work!

Who would have thought that Hollywood-conceived fiction from 1966 called Fantastic Voyage could actually become reality in our lifetime?!

Conclusion

So, in conclusion, if we go back to the original ITU specifications and see what 5G was attempting to do, hopefully, you now have some sense of the breadth of technologies that 5G will bring to humanity. As I started out by saying in part 1 of this series, 5G is really aimed at a much broader set of target customers and businesses, some of which we haven't even begun to think of and be able to conceive. Some examples that this article talks about are just starting to scratch the surface.

At the beginning of this article, I said that 5G brings together three sets of technologies, namely eMBB (High Data Rates), uRLLC (Low Latency), and mMTC (Low Power).

Figure 5

As shown in this graphic on a blog by Wave Computing, recaptured as Figure 5 above, each of these technology hubs stands independent from each other, and can provide valuable services and business opportunities in and of itself. With the current state of technology it may not be possible to combine these technology islands into one single application; For example, it is not imagined that we would be able to send 20 Gbps data within 1 millisecond at extremely low power over large distances. At least Not Yet!

But that is what we may have to contend with in the future. The term killer-app seems to have gone out of vogue in today's lexicon, but I for one, can't wait to find what those are in the future.

Onward and Upward!