5G: Hype vs Reality

Telecom companies, their suppliers, and politicians are putting 5G in the news

There have been a lot of news stories about 5G, a new mobile wireless standard. The theme many of these suspiciously similar articles is that 5G is going to transform everything. I'll tell you what to expect in reality, and what is wishful thinking on the part of the telecom industry, and why telecom service providers and equipment makers are hyping fantasies.

5G is a better radio

5G means better mobile devices and a better mobile network. There are three main reasons 5G is better:
  • 5G introduces a new radio technology that makes more efficient use of radio spectrum
  • The network behind those radios will be faster and have lower latency
  • 5G enables using more of the radio spectrum
There are many factors in the increased sophistication in 5G radios. These are the most important:
  • Encoding the digital data more densely into the radio signal
  • Transmitting and receiving signals simultaneously
  • A wider range of strategies for encoding data
  • Using multiple antennas for input and output (MIMO)
  • Forming and steering beams of radio entergy
  • Using more radio spectrum, if it's available, for an individual user's data
The 5G radio is an impressive feat of technology. Sometimes you will see 5G referred to as "5G NR." The "NR" stands for New Radio.

In addition to increased sophistication in radio technology, 5G makes use of radio bands above 6 gigahertz, also referred to as millimeter wave bands. 5G relies on these high frequency bands to provide very high speed links, in the gigabits per second.

Gigabits per second sounds pretty great, but millimeter wave radio bands have limitations: radio waves in this range can't penetrate walls, or even some windows. They need a line-of-sight between the sending and receiving antennas. They go less far when it rains.

The ability to form and steer beams of radio energy, called "beamforming," can enhance the ability to use millimeter wave bands, but using this capability effectively is challenging.

5G capabilities require a lot of computing power. Simultaneous receiving and sending requires twice the radio hardware and computing power. Beamforming requires a lot of computing, too. So does fitting more bits into the same amount of spectrum, and using more spectrum.

The economics of putting increasingly complex digital radio technology into mobile devices is favorable because billions of people buy and use mobile devices. 5G radios will be significantly more expensive at first. This year, the first 5G phones will be more expensive by hundreds of dollars. They will be thicker and heavier. Their battery life will be poorer. But, as more people buy 5G devices, the cost of making 5G chips, and the incredibly complex 5G software and chip designs, is destined to decline. The chips will get more efficient and less expensive. But that will happen over the course of several years. There is no substitute for time when refining chip designs.

Over the next two or three product generations of 5G chips and the phones that use those chips, things will be back to normal and 5G phones will cost just a little more than 4G phones and battery life will also be back to normal.

If this were the whole story the result would be that data on your mobile device will be somewhere between the same and ten times faster, and occasionally, in some locations, 100 times faster. But 5G is not just a sophisticated radio.

5G is a better, faster, and expensive network

The design of the network of radios and their connections to network nodes that run that radio network and connect it to the internet is profoundly ambitious, in multiple dimensions:
  • The physical infrastructure of the 5G radio network, when it operates in millimeter wave bands connecting to phones, requires a vast number of radios, referred to as "small cells."
  • The performance goals of the 5G network are difficult to attain and sustain
  • 5G infrastructure will be very expensive and complex to build
  • Some of the technical wizardry of 5G will be hard to do reliably in real-world networks
  • A 5G network will cost about twice as much to run and maintain as a 4G network
The advantage of a large number of small cellular radios is that some difficult scenarios can be addressed. 5G networks in airports, stadiums, and conference halls will be able to offer enough capacity for people in large crowds to get mobile wireless connections far better than they do now. These are the places where millimeter wave bands, and lots of small cells, transform the ability to solve network engineering problems.

The question is: Are there enough plausible scenarios to make wide deployment of small cells viable? As press demos of 5G have noticed, 5G millimeter wave connections work best when standing still near a 5G cell site.

5G network operators can build 5G networks gradually. 5G telecom industry standards provide more than ten different approaches to designing and building a 5G network, divided into two broad categories: Standalone (SA), where a pure purpose-built 5G network connects 5G devices, and Non-standalone (NSA) networks where the 5G New Radio is used with mostly 4G network infrastructure, alongside 4G radios. The availability of gradual paths to 5G means that, at first, on most networks, in most places, the 5G experience will not feel different from 4G.

The fully realized vision of a 5G SA network that has extremely low end-to-end latency, and that can do exotic things like "network slicing" requires an incredibly expensive reconstruction of the mobile network to fiber optic backhaul and high performance network nodes, plus extremely complex control software.

Network slicing refers to the ability to run multiple virtual networks with different performance characteristics on one set of network hardware. The use cases envisioned for network slicing include providing a "slice" for public safety users.

The risk 5G network operators who take a more aggressive approach to building out a 5G network face is that the new sources of revenue  that 5G is supposed to enable, like public safety agencies renting a network slice, 5G in factories, telemedicine, and IoT device makers using 5G extensively, don't actually materialize.

There will be islands of 5G nirvana if only because most network operators want to showcase the highest performance capabilities of 5G and test whether a complete 5G implementation adds up to a qualitative difference in the mobile user experience and in the kinds of products they can sell. Most of the planet will be 4G for a long time to come.

In the worst cases, 5G has the potential to be an outright trainwreck. For example, using millimeter wave bands for cars on the go is challenging because even car windows can block millimeter wave signals. That's right: 5G implies a mobile network that has trouble with signals penetrating cars. Solutions like antennas embedded in car windows have been proposed.

The capital and operating costs of 5G have driven both equipment makers and network operators to speculate on sometimes fanciful uses of 5G, and to reach for markets that may be mirages. In reality, many aspects of the 5G marketing vision will never happen.

Parts of 5G are a fantasy

When you specify the capabilities of telecom network equipment and write standards for those networks you create scenarios called "use cases." In 5G, this process of envisioning scenarios for how the network could be used includes things that are, to put it kindly, speculative.

Here are just some of the unlikely futures embodied in 5G use cases:
  • 5G is a critical enabling technology for autonomous vehicles
  • 5G is going to transform factories
  • 5G IoT devices will be everywhere 
  • Telemedicine is enabled by the 5G generational transition in mobile technology
  • Telcos will be in the business of providing network cache and computing resources
  • Many people will be walking around with VR or AR headgear on
Let's start with autonomous vehicles or "self-driving' cars. This is a typical "use case" that looks good if you assume the roads can be blanketed with both 5G radios and very low-latency 5G networks and network nodes. The fact is that makers of autonomous cars cannot rely on any mobile network connectivity at all, and can even less rely on or wait for pervasive low-latency 5G networks that implement features like network slicing to isolate vehicle to vehicle traffic on a dedicated set of network resources. Autonomous cars, and vehicle to vehicle (V2V) communications, are going to happen with or without 5G.

Factories and telemedicine, similarly, could use 5G radio technology, but why would they? Why would a telemedicine application make use of wireless connections when a wired connection is available? Why would a factory use wireless to gain a small increment in flexibility when moving and reprogramming factory machines makes up the vast majority of the cost of reconfiguring a production line. If you sense a bit of déjà vu, it is because telemedicine was trotted out as a key use case for 3G and 4G, and it still isn't.

One of the most speculative use cases in 5G hype is augmented reality (AR) and virtual reality (VR). Imagine tens of millions of AR users walking around with headgear on, relying on low latency to prevent the motion sickness that comes from the feeling that one's vision doesn't quite correspond to one's movement. This requires "edge computing."

AR and VR are both nascent technologies and are unlikely to be either enabled or blocked by the pace of 5G. It is also an open question whether they are going to break out of niche applications. We are unlikely to have to worry about people randomly vomiting due to lag, as they move through "mixed reality," on a regular basis.

The theory behind "edge computing" in 5G is that applications will have to be pushed out into the mobile network operators' networks to satisfy latency requirements, and that network operators will develop a huge business in assuring that applications like AR, and media like Netflix shows, get to users' phones in the optimal way.

Mobile network operators do not have the expertise that Google, Amazon, Apple, and Microsoft have in operating flexible computing resources. The implication that 5G will turn mobile network operators into the new platform for content delivery and software-as-a-service is unlikely to happen.

Seeing the world through telecom-colored glasses is not new. When 3G was launched, all sorts of fanciful scenarios were part of the public relations push. to promote the idea that 3G capabilities were transformational. But Steve Jobs and the iPhone prevailed in shaping the role of 3G: The mobile experience is the internet experience, and the role of network operators is to deliver the bits and get out of the way. So it will be with 5G.

Parts of 5G are bad

5G is good. 5G is massively over-hyped. But is it actually harmful?

Unfortunately there are aspects of 5G that are outright bad:
  • Deep packet inspection is designed-in
  • "Edge computing" means "not neutral"
  • 5G positions incumbent network operators as "too big to fail"
  • 5G is portrayed as a strategic issue critical to national security
The ability of 5G networks to guarantee the ability to deliver large critical flows in limited bandwidth and the ability to deliver low latency depends on the ability to discriminate between traffic. Some of this will be pre-configured for, for example, emergency communications. But, in many cases, 5G features depend on deep packet inspection. This means the network is looking at the content of your traffic and deciding how to treat it. This compromises end to end security and, obviously, privacy.

Deep packet inspection is inimical to end to end security, and vica versa, which makes it hostile to network neutrality. In a neutral network, no applications are favored over others. Network neutrality is key to innovation: The low cost of launching a new internet application using on-demand computing and storage resources from providers like Amazon has created an explosion in innovation. Having to negotiate with a mobile network operator for transport priority, storage, and computing resources in a 5G "edge computing" architecture makes launching apps much slower and more expensive.

Not the least harm is the impact of 5G capital requirements on a highly indebted industry. 5G's requirements for the number of radios, how they are connected, and the fiber network and new network nodes required to build the complete vision of 5G puts network operators on a collision course with property owners, local governments, and bond markets.

The two largest US mobile network operators, AT&T and Verizon roughly doubled their long term debt from about $60 billion, each, in 2010 to about $120 billion, each, in 2017. In the same period, operating cash flow remained roughly flat for AT&T, and declined for Verizon. (Source: lightreading.com)

The strains are showing: Among leading industrial nations, although the US has excellent 4G availability, the US ranks poorly for network speed and latency. (source: opensignal.com) Network performance is strongly linked to capital spending on networks. How can we expect carriers to deliver orders of magnitude better performance in 5G if they do so poorly building and operating their 4G networks?

This is one reason why part of the 5G hype is that 5G is a national strategic priority, and that China, the bogeyman of the day, will overtake the US if the US does not embark on a crash program of 5G upgrades, with government money and giveaways like free access to municipal resources.

In fact, China's own network operators are cautious. Yang Jie, chairman of China Mobile recently stated:
"Capital expenditure, including 5G, will be lower than last year's total amount."
China Unicom will...
"tighten the purse strings on 5G as it requires huge amount of investment."
...said Chairman and CEO Wang Xiaochu.

These are not the words of economic soldiers in a command economy marching lockstep toward 5G domination. But that hasn't stopped analogies to a "missile gap." In an opinion piece in Newsweek Newt Gingrich, writes:
“Go” is an ancient Chinese board game based on encirclement and territorial control. It is the most ancient and complicated board game in the world. Beijing is engaged in a concerted strategy of encirclement and control of wireless. But too many nations in the West are content to “let the chips fall where they may.”
To be fair, it's good to see Newt has let go of free market dogma. He proposes a remedy to 5G problems: a government funded nationwide 5G network:
The project should be nationwide, with broad geographic coverage—in contrast to current operators’ plans for targeted, urban-specific 5G rollouts, which leave rural America in a 3G or 4G world. This will benefit those on the wrong side of the digital divide while making possible a wider range of innovative uses of the network. These include precision agriculture, automotive and trucking telemetry, telemedicine, and many other advancements.
This closes the circle perfectly. All the fantasy use cases get implemented, and mobile network operators' debt doesn't increase. The peril from China is held at bay. America strides forward into the 5G future.

5G will be mostly the same

When it comes to 5G, don't fear it. It will make your phone faster. It will make the mobile network better. More data will fit in the same spectrum. Eventually, it won't also make your phone hotter and run out of battery.

Don't buy in to the hype. Guard your wallet from people claiming 5G is a strategic imperative. Be vigilant against network operators leveraging 5G to capture what are currently open and competitive elements of the internet infrastructure.

5G will be there, when and where it makes economic sense. So will WiFi 6, but that is another story.


Comments

  1. Very simply, 5G is an attempt to return to the 1970's intelligent network and claw value back from the Internet. It also means a return of the busy signal and scarcity. It creates brittle dependency on network operators (and their fees).

    Basically using the fancy radio to frame the larger story taking real engineering and using it to justify a business agenda. Like paying to demonize fat while saying sugar is wonderful.

    ReplyDelete

Post a Comment

Popular Posts

The QUIC Brown Fox Jumped Over the Top of Carrier Messaging, or Allo, Duo, WebRTC, QUIC, Jibe, and RCS, Explained

Telirati Analysis #11 Diagnosing and Fixing Google's Social Problems