The Network Superhighway — How 5G will change the world
In my last post, I talked about the road to 5G and how each successive network standard has paved the way to where we stand today. From the audio-only first generation (1G) network to the packet-switched 4G LTE standard — we’re now speeding towards an infrastructure that can tackle the needs of millions of interconnected devices, all demanding real-time interactions. You’ve heard and read a lot about 5G. Here’s a deeper dive into what 5G is, its potential, and also its constraints.
Some 5G use cases
Whether you like it or not, we’re living in the age of data. Almost all current (and future) technologies — from Alexa to self-driving vehicles — need connectivity to transmit this data, connectivity that combines high, almost real-time speed with low latency.
Here are just a few examples of innovations that require this upcoming level of connectivity:
- Multiplayer gaming which depends on split-second responses (and is already a multi-billion dollar industry)
- Crash avoidance systems for cars and public transport
- Autonomous parking that works with close interaction between the automobile (cars, buses, trains) as well as the building
- Advanced patient monitoring through sensors on wearable technologies
- Real-time clinical decisions and remote surgery
- High-resolution video streaming — which becomes even more important when 4k and 8k resolution screens become commonplace.
These are the areas where 5G truly shines.
Given all the buzz surrounding 5G, every company seems to be trying to make waves with the new technology. Major U.S. telecommunications companies such as Verizon and AT&T have already announced plans to launch 5G service by early 2019. Ericsson is building out 5G as a global standard across all continents. And other notable market leaders in the space are Nokia, ZTE, Qualcomm, and Samsung. My company DASAN Zhone Solutions recently announced a 3-year partnership with LG U+ to supply 5G solutions in South Korea.
Chinese telecom behemoth Huawei owns a significant 28% market share in the industry and have been at the forefront of 5G R&D, but they are currently embroiled in an international scandal which I wrote about previously, starting from the arrest of its CFO in early December 2018 and most recently the filing of serious indictments by the US Justice Department at the end of January 2019.
Despite this fervent press coverage in a seemingly crowded space, one less often discussed fact is the high cost of this upgrade. In order to provide a network that is robust enough to meet consumer and business demand, players in this space can expect a combined research and capital spending outlay of an estimated $200 billion a year. From an infrastructure perspective, a 5G upgrade means significantly more cell towers, each interconnected with expensive fiber optic cables.
How does 5G work and why is it so expensive?
Latency is defined as the time it takes for a packet of data to move from a source to a receiver. 4G LTE, while fast, has a latency of 80–100 milliseconds. 5G, on the other hand, shaves latency down to 1–2 milliseconds — almost 100x better.
These higher speeds are achieved by utilizing a high-energy bandwidth with greater frequency which, in turn, allows the network to carry much higher amounts of data. However, since the energy loss is also proportional, this causes a significant reduction in the range. The way to counteract this is through a denser, fiber-rich network infrastructure. The limited range drives the need for a larger number of small cells, all interconnected by fiber optic cables.
Essentially, fiber optic cables are glass/plastic wires with refractive inner surfaces. Light travels along these wires and transmits data from one end to another.
Fiber optics work on the principle of total internal reflection, ensuring that data loss is negligible. Characteristics of this material include:
- Greater bandwidth, which allows it to carry more data
- Lower susceptibility to interference, due to minimal resistance and no interference from magnetic waves
- Digital as opposed to analog data transfer
- Much thinner and lighter than metal wiring
- Much longer lifecycle
However, these cables are much more fragile than their metallic counterparts, reducing the flexibility to splice one and connect it to another using a regular wire stripper. They are also more expensive to install.
Currently, the backbone of our mobile infrastructure which connects mobile switching stations, is already using fiber optics for wired data transfer. This part of the infrastructure, known as the ‘backhaul’, is essentially responsible for connecting each call to the right phone.
With 5G, the amount of data being transferred in the ‘fronthaul’ — that is, the part of the network responsible for maintaining stability when a person moves between cell towers — increases significantly. To ensure that the entire network responds with latency and capacity that 5G promises, this section of the network also needs to be upgraded to fiber.
This ‘last mile connectivity’ is a colossal investment at the data transmitter/receiver level. To put that into perspective, the state of California alone has over thirteen thousand stations that would need to be upgraded. This, paired with 5G’s shorter range, might require cell towers on every street corner.
And do we really *need* 5G right now?
We’ve spent a bunch of this post talking about the supply side of 5G; let’s also not neglect to cover the demand side. In most developed geographies like the US and Europe, areas where 5G is most likely to be technologically useful, the industry is already highly saturated and acquiring new subscribers is difficult without competing on price.
“Historically, 1G to 4G, it’s been a pretty straightforward evolution from the point of view of business and technology. The revenue grew proportionate to the usage” said wireless expert Chetan Sharma. “This time around, it’s not clear that 5G will translate into more revenue until perhaps five or 10 years from now.”
5G-enabled applications like the ones I listed in the first section are still in R&D infancy and may take years to come to market, much less return much profit. And any consumer mobile upgrade to 5G would necessitate purchasing a new mobile device, a proposition that may be tough if the clear benefits can’t be cited. Indeed, it’s not surprising to see companies like Apple delay 5G adoption until 2020 in order to allow for the technological kinks to be worked out and for there to be more widespread availability.
A study by Deloitte predicts that by the end of 2020, 5G mobiles will account for 15–20 million devices. Sales are expected to cross 100 million by 2021, though that number is still under 5% of total industry sales.
5G is coming, but I believe it won’t be here as quickly as the media and the manufacturers would like us to think. There are clear PR benefits to jumping on the 5G bandwagon right now, so companies are eager to publicize their entry to 5G even though the economics don’t currently make sense. It will take at least 3–5 years before this technology becomes feasible, commercially viable, and commonly used — and the giants, both countries and conglomerates, that are investing in this infrastructure are well aware of that.
5G is the future, but that future is still quite a ways away.