2G, 3G, 4G LTE Network Shutdown Updates

By Harald Remmert, Sr. Director of Engineering Digi

With some cellular networks sunsetting and 5G networks growing, the entire landscape of cellular connectivity is changing. For businesses that have deployed devices based on older networks, the questions today are around the life expectancy of those networks, how long 4G LTE networks will be viable, and if 5G will mature soon enough for their needs. This article will further explore and answer these questions.

Why older networks shut down

Carriers need to be able to re-use spectrum to grow new networks and provide faster, more responsive technology for their customers. Old 2G/3G infrastructure must make way for new networks, and this means older cellular devices will no longer be viable and must be retired.

There are two metrics in this context that matter most:

  1. Spectral efficiency, measured in bits per second/Hertz (Hz). This is an indicator of how efficient the data is transferred per available bandwidth. Modulation, coding schemes and error correction play a key role here. Higher order modulation schemes like 64-QAM and 256-QAM commonly used in 4G LTE allow data to transfer up to 8 times more bits/Hz as compared to 2G or 3G.
  2. Latency. This is the delay from request to response, and is an indicator of how responsive a cellular network is. Through cellular network and technology upgrades, latency has come down from seconds in the early 2G days, and triple-digit milliseconds in 3G, to double-digit milliseconds in 4G LTE networks and an anticipated single-digit latency in future 5G Stand-Alone (SA) networks.

The good news is that 4G LTE will be available for at least a decade to come, and will co-exist with 5G networks. This article will provide updates on the sunsetting of 2G and 3G networks and the outlook for 4G LTE and 5G networks to support those who are planning their migration path.

The outlook for 2G sunsetting

Image of 2G sunsettingFigure 1: All major North American carriers will be phasing out 2G by the end of 2021. (Image source: Digi)

In the U.S., large carriers are already in the process of shutting down 2G service:

  • AT&T stopped servicing 2G networks in 2016
  • T-Mobile will roll up the sidewalks on 2G in December 2020
  • Verizon Wireless will also phase out its 2G CDMA network at the end of 2020
  • Sprint will shut down their 2G CDMA network in December 2021

However, before these 2G networks are fully decommissioned, 2G devices will stop performing optimally. Here’s what to expect:

  1. 2G cellular devices may still connect to the network, but it will not be possible to activate new devices.
  2. Those devices will likely not work as well as in the past, as carriers will be re-farming that spectrum to make room for new networks coming online.

Elsewhere in the world:

  • Bell Canada closed down 2G networks in 2018
  • Other carriers in Canada include Telus and Rogers, which have both stopped supporting 2G
  • In Europe, 2G will have a longer lifeline. Vodafone intends to support 2G service until at least 2025. However, note that they will likely begin re-farming spectrum prior to that time.

The outlook for 3G sunsetting

Image of 3G sunsettingFigure 2: Major North American carriers plan to phase out 3G by the end of 2022. (Image source: Digi)

Like 2G networks, 3G networks use spectrum that will ultimately be needed to support 4G LTE, which means carriers are planning phase-outs. Additionally, 3G devices have speed limitations, achieving around 3 Mbps, at the maximum. By contrast, 4G LTE is both faster and uses spectrum more efficiently, which means more devices can share available spectrum.

Current plans from major carriers are as follows:

  • Verizon will shut down its CDMA network at the end of 2020
  • AT&T plans to shut down their 3G network in February of 2022 and 3G phones can no longer be activated
  • T-Mobile is expected to sunset their 3G networks by the end of 2021
  • Sprint 3G networks will be shut down in December of 2022, and 3G devices can no longer be activated

This is a very good time to evaluate 4G LTE device deployments, which will be discussed next.

The outlook for 4G LTE networks

Image of no plans to phase out 4GFigure 3: While 2G and 3G will be phased out soon, there are no plans to phase out 4G in the immediate future. (Image source: Digi)

While 4G LTE will eventually become obsolete, no one should hold their breath until the shutdown, which is a decade or more in the future. Enterprises planning IoT deployments today can safely plan on long-term viability of LTE networks.

In fact, not only will it be at least a decade or more before 5G completely overtakes 4G, but 4G LTE provides all of the bandwidth most applications need today, and is a far more cost-effective choice. 4G LTE, which stands for Long Term Evolution, will very effectively share spectrum with 5G networks as they grow, with many devices over time being developed with both 4G and 5G functionality.

Today, 5G is primarily in use by early adopters who are willing to pay a high premium – either for the novelty of it or to perform early, limited trials. As with all new technology, important incremental development must occur to enable widespread adoption.

Key considerations for 5G adoption

Here are the some important considerations to help with migration planning of current systems:

  • 5G networks are cutting edge, but coverage is limited today, and network infrastructure must evolve to deliver the 5G vision of “one ubiquitous network for all use cases”.
  • Technologies like 5G mmWave are not ideal in areas with obstructions, such as buildings or dense foliage, or locations that experience a lot of precipitation or snow. Even double-paned glass can block a 5G mmWave signal.
  • Due to the need for development and testing to ensure signal viability, as well as node density, the full 5G rollout will occur over the next 3-5 years and longer, depending on the geographic area.

Diagram of 5G will enable faster data rates, larger device density, and lower latencyFigure 4: 5G will enable faster data rates, larger device density, and lower latency. (Image source: Digi)

  • Like 4G LTE, 5G is about taking a long-term evolution approach and providing incremental value along the way. This can best be seen in the specifications behind the technology, which are bundled into releases by the 3rd Generation Partnership Project (3GPP):
    • 3GPP Release 15 was the first 5G release in December 2018, focusing on the foundation and Enhanced Mobile Broadband (which means high-speed applications).
    • Release 16 (sometimes referred to as 5G Phase 2) was released in July 2020. It focused on mission-critical control, but also evolved Enhanced Mobile Broadband.
    • Release 17 is scheduled for late 2021/early 2022, and will focus on the Massive Internet of Things, and evolve the other two areas as well, through lower latency and even 5G over satellites.

How 4G LTE and 5G will share the road

While carriers needed to shut down 2G and 3G networks to accommodate 4G LTE, this is not the case with 4G and 5G. 5G uses new spectrum, such as mmWave, and it can work in conjunction with the 4G existing spectrum. This is due to Dynamic Spectrum Sharing (DSS).

What is Dynamic Spectrum Sharing (DSS)?

DSS is a technology that allows the deployment of both 4G LTE and 5G in the same frequency band. It dynamically allocates spectrum resources between the two technologies, based on user demand.

Without DSS, an operator who has 20 MHz of 4G LTE spectrum available would have to split that spectrum, meaning they would have to allocate 10 MHz of spectrum to 4G LTE and cram all their LTE users into that 10 MHz of spectrum. Then, the remaining 10 MHz spectrum could be used for 5G, even though initially there will only be a minimal number of 5G users.

With DSS, an operator doesn’t have to split their spectrum or have a dedicated spectrum for either 4G LTE or 5G. Instead, they can share that 20 MHz of spectrum between the two technologies. Initially, the spectrum will be mostly used for 4G LTE, but as the number of 5G devices grows, more and more spectrum will be used for 5G. Eventually, in an estimated 15 to 20 years when there are few 4G LTE devices in use, most if not all of the spectrum will be used for 5G.

Dynamic spectrum sharing lets carriers use the same spectrum band for both 4G and 5G. To use a traffic analogy, instead of building different roads for buses and cars, DSS is like having one big highway with separate lanes for the different types of vehicles.

From 5G non-standalone to 5G standalone

Many 5G devices, like cellular routers, will be built with both 4G and 5G radios so that they’re capable of connecting to either network. In the early days of 5G, they must connect to 4G LTE first, and then connect to a 5G network if it’s available. This is also called 5G Non-Standalone (NSA).

At first, most data will be transmitted via 4G LTE. However, over time as 5G coverage expands and the cellular infrastructure moves to 5G Stand-alone (SA), more and more data will be transmitted via 5G and cellular devices can connect to the 5G network directly without the detour over 4G LTE. Therefore, to continue with the traffic analogy, 5G will eventually overtake 4G as the bigger lane on the freeway.

How do LTE-M and NB-IoT fit into 5G?

Suppose deploying a larger installation of LTE-M or NB-IoT devices is desired in the near future. Will these devices soon be obsolete, as 5G becomes available? Should there be a delay in deployment while waiting for 5G?

The answer is that a great technology choice was made with LTE-M or NB-IoT, because while these are 4G LTE technologies, they were developed as part of LTE’s long-term evolution paradigm, with 5G in mind.

When LTE-M and NB-IoT were initially designed, special attention was given to the design of 5G to make certain LTE-M and NB-IoT can operate or co-exists in-band to a 5G system. This provides a forward-compatible path well into the 5G future, beyond when 4G LTE may no longer be available.

As mentioned earlier, Release 17, scheduled for late 2021/early 2022, will be the first release focusing on the Massive Internet of Things, which means that chipsets and devices will follow with the earliest being available in 2022/2023.

Four ways to prepare for 5G today

What can be done today to prepare for 5G? Should an upgrade be considered now or wait? Is 5G really needed today? Digi is currently hearing these and many other questions from customers. In response to these questions, the following actionable process was put together:

1. Identify 5G applications

Identify applications where much faster speed and lower latency would optimize operations. A prime example are applications that leverage edge compute for machine learning and predictive maintenance.

Will application needs change over the next 5 years in terms of data volume, latency, and power requirements? For example, LTE-M and NB-IoT enable new battery-operated devices and new business models that were previously not possible with 2G/3G.

Additional questions to ask: What flavor of 5G is needed? Is a 5G sub-6 with its nationwide coverage needed but with similar performance to 4G LTE today, or is the high-speed of 5G mmWave, which is mostly available in densely populated, urban environments needed? This is an important factor, because as mentioned earlier, not all spectrum will be available everywhere, and spectrum availability will evolve over time. Is it a desire to 5G-enable or replace existing equipment?

2. Take stock of device inventory

Understanding the appropriate 4G to 5G evolution path will help in planning for the future. Perhaps 2G or 3G devices are still in use or the application is using first-generation 4G devices which may benefit from an upgrade to newer 4G LTE devices with faster processors and 4G Gigabit LTE cellular speeds.  If this is the case, look for modular devices or devices that are “5G Ready”, which typically means that they have the horsepower and interfaces to support 5G via a radio upgrade.

3. Quantify the cost or risk of maintaining the current deployment

If 2G or 3G devices are still in use, there is a risk of connectivity loss and service interruption due to network shutdowns. If using 4G LTE devices, these are generally considered good, but look at the age of the device and determine if it is desired to replace 1:1 or upgrade to a newer device. Key factors to consider are device security and firmware updates. As always, Digi can be a guide in these choices.

4. Create a business case for migration

Do a cost/benefit analysis that shows when and if moving to 5G is a worthwhile move. Look at meaningful interim technology steps as part of a larger migration. For example, consider a modular, higher-end 4G LTE/5G-ready device now that also provides an upgrade path when 5G becomes available and the applicable business case can support a migration to 5G.

Work with the device and mobile network operator partners to align on device and network availability timelines. Consider installation costs, such as for outdoor 5G mmWave equipment.

Planning for change

Today, especially in developed countries, those preparing for device deployments have a great choice in 4G LTE, which has plenty of speed and reasonably low latency for most applications.

In less developed countries, 4G LTE will likely remain the standard for several decades to come.

At the same time, 5G network coverage will grow, and the technology will ultimately deliver the ultra-fast speed and low latency that is anticipated to be game changers for advanced technologies such as autonomous vehicles.

In the case of 2G/3G device deployments, the days of their viability are limited and it is time to plan by beginning or continuing 4G deployments with confidence. Being an early adopter of 5G in the commercial or industrial space is being a pioneer. Reach out to Digi to discuss future plans and how to get support for application requirements.

Disclaimer: The opinions, beliefs, and viewpoints expressed by the various authors and/or forum participants on this website do not necessarily reflect the opinions, beliefs, and viewpoints of Digi-Key Electronics or official policies of Digi-Key Electronics.

About this author

Harald Remmert, Sr. Director of Engineering Digi

Harald Remmert is an accomplished technology leader in the IoT ecosystem with over 20 years of experience in product strategy, design, development, testing and engineering leadership with a proven track-record of introducing successful products to the market. Harald has deep technical knowledge in mixed hardware and software product development. In his current role at Digi he’s on the lookout for new technologies and tools such as 5G, Edge Compute, Machine Learning & Artificial Intelligence to innovate and solve business problems more efficiently. Harald has deep technical knowledge in mixed hardware and software product development, and he’s always on the lookout for new technologies and tools to innovate and solve business problems more efficiently.